SPECIAl ED.
NY PUBLIC LIBRARY THE Bf"^f|':^|}-!^[jf^I|m|
3 3333 05990 5162
Digitized by the Internet Archive
in 2007 with funding from
Microsoft Corporation
http://www.archive.org/details/fabresbookofinseOOfabr
FABRE'S BOOK OF INSECTS
>*i!'
Fabre
FAB RE'S
BOOK OF INSECTS
RETOLD FROM ALEXANDER TEDCEIRADE MATTOS'
TRANSLATION of FABRE'S "SOUVENIRS ENTOMOLOGIQUES"
BY MRS.RODOLPH STAWELL,
Illustrated Ly
EJDETMOLD
NEW YORK
DODD, MEAD AND COMPANY
1926
COPTKIGHT, 1921,
BY DODD. MEAD AND COMPANY, iNtt
IONS. J
PBINTED IN U. S. A.
T
CONTENTS
CHAPTER I
PAGE
MY WORK AND MY WORKSHOP i
CHAPTER II
THE SACRED BEETLE ii
CHAPTER III
THE CICADA 25
CHAPTER IV
THE PRAYING MANTIS 40
CtlAPTER V ,; ***
THE GLOW-WORM . /'i^'^J;;. .... 54
CHAPTER VI
A MASON-WASP ^
CHAPTER VII
THE PSYCHES ^9
vii
R 8258
CONTENTS
CHAPTER VIII
PAGE
THE SELF-DENIAL OF THE SPANISH COPRIS . .109
CHAPTER IX
TWO STRANGE GRASSHOPPERS 121
CHAPTER X
COMMON WASPS 138
CHAPTER XI
THE ADVENTURES OF A GRUB 157
CHAPTER XII
THE CRICKET 175
CHAPTER XIII
THE SISYPHUS 198
" ..CHAPTER XlV
THE CAPRICORN .^;^^r.^^ 209
CHAPTER XV
LOCUSTS , 227
CHAPTER XVI
THE ANTHRAX FLY 249
viii
LIST OF ILLUSTRATIONS
THE SACRED BEETLE . . . Fronttsptece
Sometimes the Scarab seems to enter into partnership with a friend
THE CICADA
FACING
In July, when most of the insects in my sunny country are parched with
thirst, the Cicada remains perfectly cheerful 26
THE PRAYING MANTIS
A long time ago, in the days of ancient Greece, this insect was named
Mantis, or the Prophet ^^2
PELOP^US SPIRIFEX
When finished the work is amber-yellow, and rather reminds one of the
outer skin of an onion 80
THE PSYCHES
This is the secret of the walking bundle of sticks. It is a Faggot
Caterpillar, belonging to the group known as the Psyches ... 90
THE SPANISH COPRIS
The burrow is almost filled by three or four ovoid nests, standing one
against the other, with the pointed end upwards 116
THE WHITE-FACED DECTICUS
The Greek word dectikos means biting, fond of biting. The Decticus
is well named. It is eminently an insect given to biting . . . .130
COMMON WASPS
The wasp's nest is made of a thin, flexible material like brown paper,
formed of particles of wood ^44
THE FIELD CRICKET
Here is one of the humblest of creatures able to lodge himself to
perfection. He has a home; he has a peaceful retreat, the first
condition of comfort ^
LIST OF ILLUSTRATIONS
THE SISYPHUS
PAQB
The mother harnesses herself in the place of honour, in front. The
father pushes behind in the reverse position, head downwards . . 204
ITALIAN LOCUSTS
"I have buried underground," she says, "the treasure of the future" . . 238
THE ANTHRAX FLY
Her delicate suit of downy velvet, from which you take the bloom by
merely breathing on it, could not withstand the contact of rough
tunnels 258
FABRE'S BOOK OF INSECTS
FABRE'S BOOK OF INSECTS
CHAPTER I
MY WORK AND MY WORKSHOP
WE all have our own talents, our special gifts.
Sometinies these gifts seem to come to us
from our forefathers, but more often it is
difficult to trace their origin.
A goatherd, perhaps, amuses himself by counting little
pebbles and doing sums with them. He becomes an as-
toundingly quick reckoner, and in the end is a professor
of mathematics. Another boy, at an age when most of
us care only for play, leaves his schoolfellows at their
games and listens to the imaginary sounds of an organ, a
secret concert heard by him alone. He has a genius for
music. A third — so small, perhaps, that he cannot eat
his bread and jam without smearing his face — takes a
keen delight in fashioning clay into little figures that are
amazingly lifelike. If he be fortunate he will some day
be a famous sculptor.
To talk about oneself is hateful, I know, but perhaps
I may be allowed to do so for a moment, in order to intro-
duce myself and my studies.
[1]
FABRE'S BOOK OF INSECTS
From my earliest childhood I have felt drawn towards
the things of Nature. It would be ridiculous to suppose
that this gift, this love of observing plants and insects,
was inherited from my ancestors, who were uneducated
people of the soil and observed little but their own cows
and sheep. Of my four grandparents only one ever
opened a book, and even he was very uncertain about his
spelling. Nor do I owe anything to a scientific training.
Without masters, without guides, often without books, I
have gone forward with one aim always before me: to
add a few pages to the history of insects.
As I look back — so many years back I — I can see my-
self as a tiny boy, extremely proud of my first braces and
of my attempts to learn the alphabet. And very well I
remember the delight of finding my first bird's nest and
gathering my first mushroom.
One day I was climbing a hill. At the top of it was a
row of trees that had long interested me very much.
From the little window at home I could see them against
the sky, tossing before the wind or writhing madly in the
snow, and I wished to have a closer view of them. It
was a long climb — ever so long; and my legs were very
short. I clambered up slowly and tediously, for the
grassy slope was as steep as a roof.
Suddenly, at my feet, a lovely bird flew out from its
[2]
MY WORK AND MY WORKSHOP
hiding-place under a big stone. In a moment I had
found the nest, which was made of hair and fine straw,
and had six eggs laid side by side in it. The eggs were
a magnificent azure blue, very bright. This was the first
nest I ever found, the first of the many joys which the
birds were to bring me. Overpowered with pleasure, I
lay down on the grass and stared at it.
Meanwhile the mother-bird was flying about uneasily
from stone to stone, crying ''Tack! Tack!" in a voice of
the greatest anxiety. I was too small to understand what
she was suffering. I made a plan worthy of a little beast
of prey. I would carry away just one of the pretty blue
eggs as a trophy, and then, in a fortnight, I would come
back and take the tiny birds before they could fly away.
Fortunately, as I walked carefully home, carrying my blue
egg on a bed of moss, I met the priest.
"Ah!" said he. "A Saxicola's egg! Where did you
getitr'
I told him the whole story. "I shall go back for the
others," I said, 'when the young birds have got their
quill-feathers."
"Oh, but you mustn't do that!" cried the priest.
"You mustn't be so cruel as to rob the poor mother of
all her little birds. Be a good boy, now, and promise not
to touch the nest."
[3]
FABRE'S BOOK OF INSECTS
From this conversation I learnt two things: first,
that robbing birds' nests is cruel and, secondly, that birds
and beasts have names just like ourselves.
"What are the names of all my friends in the woods
and meadows?" I asked myself. "And what does
Sdxicolu mean?" Years later I learnt that Saxicola
means an inhabitant of the rocks. My bird with the
blue eggs was a Stone-chat.
Below our village there ran a little brook, and beyond
the brook was a spinney of beeches with smooth, straight
trunks, like pillars. The ground was padded with moss.
It was in this spinney that I picked my first mushroom,
which looked, when I caught sight of it, like an egg
dropped on the rnpss by some wandering hen. There
were many others there, of different sizes, forms, and
colours. Some were shaped like bells, some like
extinguishers, some like cups: some were broken, and
were weeping tears of milk: some became blue when
I trod on them. Others, the most curious of all, were
like pears with a round hole at the top — a sort of chimney
whence a whiif of smoke escaped when I prodded their
under-side with my finger. I filled my pockets with
these, and made them smoke at my leisure, till at last
they were reduced to a kind of tinder.
Many a time I returned to that delightful spinney,
[4]
MY WORK AND MY WORKSHOP
and learnt my first lessons in mushroom-lore in the
company of the Crows. My collections, I need hardly
say, were not admitted to the house.
In this way — by observing Nature and making experi-
ments— nearly all my lessons have been learnt: all
except two, in fact. I have received from others two
lessons of a scientific character, and two only, in the
whole course of my life: one in anatomy and one in
chemistry.
I owe the first to the learned naturalist Moquin-Tan-
don, who showed me how to explore the interior of a
Snail in a plate filled with water. The lesson was short
and fruitful.^
My first introduction to chemistry was less fortunate.
It ended in the bursting of a glass vessel, with the result
that most of my fellow-pupils were hurt, one of them
nearly lost his sight, the lecturer's clothes were burnt to
pieces, and the wall of the lecture-room was splashed
with stains. Later on, when I returned to that room, no
longer as a pupil but as a master, the splashes were still
there. On that occasion I learnt one thing at least.
Ever after, when I made experiments of that kind, I kept
my pupils at a distance.
It has always been my great desire to have a laboratory
1 See Insect Adventures, retold for young people from the works of Henri Fabre.
[5]
FABRE'S BOOK OF INSECTS
in the open fields — not an easy thing to obtain when one
lives in a state of constant anxiety about one's daily
bread. For forty years it was my dream to own a little
bit of land, fenced in for the sake of privacy: a
desolate, barren, sun-scorched bit of land, overgrown
with thistles and much beloved by Wasps and Bees.
Here, without fear of interruption, I might question the
Hunting-wasps and others of my friends in that difficult
language which consists of experiments and observa-
tions. Here, without the long expeditions and rambles
that use up my time and strength, I might watch my
insects at every hour of the day.
And then, at last, my wish was fulfilled. I obtained a
bit of land in the solitude of a little village. It was a
harmas, which is the name we give in this part of
Provence to an untilled, pebbly expanse where hardly
any plant but thyme can grow. It is too poor to be worth
the trouble of ploughing, but the sheep pass there in
spring, when it has chanced to rain and a little grass
grows up.
My own particular harmas, however, had a small
quantity of red earth mixed with the stones, and had
been roughly cultivated. I was told that vines once
grew here, and I was sorry, for the original vegetation
had been driven out by the three-pronged fork. There
[6]
MY WORK AND MY WORKSHOP
was no thyme left, nor lavender, nor a single clump of
the dwarf oak. As thyme and lavender might be useful
to me as a hunting-ground for Bees and Wasps, I was
obliged to plant them again.
There were plenty of weeds : couch-grass, and prickly
centauries, and the fierce Spanish oyster-plant, with its
spreading orange flowers and spikes strong as nails.
Above it towered the Illyrian cotton-thistle, whose
straight and solitary stalk grows sometimes to the height
of six feet and ends in large pink tufts. There were
smaller thistles too, so well armed that the plant-collector
can hardly tell where to grasp them, and spiky knap-
weeds, and in among them, in long lines provided with
hooks, the shoots of the blue dewberry creeping along
the ground. If you had visited this prickly thicket with-
out wearing high boots, you would have paid dearly for
your rashness I
Such was the Eden that I won by forty years of
desperate struggle.
This curious, barren Paradise of mine is the happy
hunting-ground of countless Bees and Wasps. Never
have I seen so large a population of insects at a single
spot. All the trades have made it their centre. Here
come hunters of every kind of game, builders in clay,
cotton-weavers, leaf-cutters, architects in pasteboard.
[7]
FABRE'S BOOK OF INSECTS
plasterers mixing mortar, carpenters boring wood, miners
digging underground galleries, workers in gold-beaters'
skin, and many more.
See— here is a Tailor-bee. She scrapes the cobwebby
stalk of the yellow-flowered centaury, and gathers a ball
of wadding which she carries off proudly with her
mandibles or jaws. She will turn it, underground, into
cotton satchels to hold the store of honey and the eggs.
And here are the Leaf-cutting Bees, carrying their black,
white, or blood-red reaping brushes under their bodies.
T\\v\ will visit the neighbouring shrubs, and there cut
from the leaves oval pieces in which to wrap their harvest.
Here too are the black, velvet-clad Mason-bees, who
work with cement and gravel. We could easily find
specimens of their masonry on the stones in the harmas.
Next comes a kind of Wild Bee who stacks her cells in the
winding staircase of an empty snail-shell; and another
who lodges her grubs in the pith of a dry bramble-stalk;
and a third who uses the channel of a cut reed; and a
fourth who lives rent-free in the vacant galleries of some
Mason-bee. There are also Bees with horns, and Bees
with brushes on their hind-legs, to be used for reaping.
While the walls of my harmas were being built some
great heaps of stones and mounds of sand were scattered
here and there by the builders, and were soon occupied
by a variety of inhabitants. The Mason-bees chose the
[8]
MY WORK AND MY WORKSHOP
chinks between the stones for their sleeping-place. The
powerful Eyed Lizard, who, when hard pressed, attacks
both man and dog, selected a cave in which to lie in wait
for the passing Scarab, or Sacred Beetle. The Black-
eared Chat, who looks like a Dominican monk in his
white-and-black raiment, sat on the top stone singing his
brief song. His nest, with the sky-blue eggs, must have
been somewhere in the heap. When the stones were
moved the little Dominican moved too. I regret him:
he would have been a charming neighbour. The Eyed
Lizard I do not regret at all.
The sand-heaps sheltered a colony of Digger-wasps
and Hunting-wasps, who were, to my sorrow, turned out
at last by the builders. But still there are hunters left:
some who flutter about in search of Caterpillars, and
one very large kind of Wasp who actually has the cour-
age to hunt the Tarantula. Many of these mighty
Spiders have their burrows in the harmas, and you can
see their eyes gleaming at the bottom of the den like
little diamonds. On hot summer afternoons you may
also see Amazon-ants, who leave their barracks in long
battalions and march far afield to hunt for slaves.
Nor are these all. The shrubs about the house are
full of birds. Warblers and Greenfinches, Sparrows and
Owls: while the pond is so popular with the Frogs that
in May it becomes a deafening orchestra. And boldest
[9]
FABRE'S BOOK OF INSECTS
of all, the Wasp has taken possession of the house itself.
On my doorway lives the White-banded Sphex: when
I go indoors I must be careful not to tread upon her as
she carries on her work of mining. Just within a closed
window a kind of Mason-wasp has made her earth-built
nest upon the freestone wall. To enter her home she
uses a little hole left by accident in the shutters. On
the mouldings of the Venetian blinds a few stray Mason-
bees build their cells. The Common Wasp and the
Solitary Wasp visit me at dinner. The object of their
visit, apparently, is to see if my grapes are ripe.
Such are my companions. My dear beasts, my friends
of former days and other more recent acquaintances, are
all here, hunting, and building, and feeding their
families. And if I wish for change the mountain is
close to me, with its tangle of arbutus, and rock-roses,
and heather, where Wasps and Bees delight to gather.
And that is why I deserted the town for the village, and
came to Serignan to weed my turnips and water my
lettuces.
[10]
CHAPTER II
THE SACRED BEETLE
I
THE BALL
IT is six or seven thousand years since the Sacred
Beetle was first talked about. The peasant of
ancient Egypt, as he watered his patch of onions
in the spring, would see from time to time a fat black
insect pass close by, hurriedly trundling a ball backwards.
He would watch the queer rolling thing in amazement,
as the peasant of Provence watches it to this day.
The early Egyptians fancied that this ball was a
symbol of the earth, and that all the Scarab's actions
were prompted by the movements of the heavenly
bodies. So much knowledge of astronomy in a Beetle
seemed to them almost divine, and that is why he is called
the Sacred Beetle. They also thought that the ball he
rolled on the ground contained the egg, and that the
young Beetle came out of it. But as a matter of fact,
it is simply his store of food.
It is not at all nice food. For the work of this Beetle
[11]
FABRE'S BOOK OF INSECTS
is to scour the filth from the surface of the soil. The ball
he rolls so carefully is made of his sweepings from the
roads and fields.
This is how he sets about it. The edge of his broad,
flat head is notched with six teeth arranged in a semi-
circle, like a sort of curved rake; and this he uses for
digging and cutting up, for throwing aside the stuff he
does not want, and scraping together the food he chooses.
His bow-shaped fore-legs are also useful tools, for they
are very strong, and they too have five teeth on the out-
side. So if a vigorous effort be needed to remove some
obstacle the Scarab makes use of his elbows, that is to
say he flings his toothed legs to right and left, and clears
a space with an energetic sweep. Then he collects arm-
fuls of the stuff he has raked together, and pushes it
beneath him, between the four hinder-legs. These are
long and slender, especially the last pair, slightly bowed
and finished with a sharp claw. The Beetle then presses
the stuff against his body with his hind-legs, curving
it and spinning it round and round till it forms a perfect
ball. In a moment a tiny pellet grows to the size of
a walnut, and soon to that of an apple. I have seen
some gluttons manufacture a ball as big as a man's fist.
When the ball of provisions is ready it must be moved
to a suitable place. The Beetle begins the journey. He
clasps the ball with his long hind-legs and walks with
[12]
THE SACRED BEETLE
his fore-legs, moving backwards with his head down and
his hind-quarters in the air. He pushes his load behind
him by alternate thrusts to right and left. One would
expect him to choose a level road, or at least a gentle in-
cline. Not at all I Let him find himself near some
steep slope, impossible to climb, and that is the very path
the obstinate creature will attempt. The ball, that enor-
mous burden, is painfully hoisted step by step, with in-
finite precautions, to a certain height, always backwards.
Then by some rash movement all this toil is wasted:
the ball rolls down, dragging the Beetle with it. Once
more the heights are climbed, and another fall is the
result. Again and again the insect begins the ascent.
The merest trifle ruins everything; a grass-root may trip
him up or a smooth bit of gravel make him slip, and
down come ball and Beetle, all mixed up together. Ten
or twenty times he will start afresh, till at last he is
successful, or else sees the hopelessness of his efforts
and resigns himself to taking the level road.
Sometimes the Scarab seems to enter into partnership
with a friend. This is the way in which it usually hap-
pens. When the Beetle's ball is ready he leaves the
crowd of workers, pushing his prize 'backwards. A
neighbour, whose own task is hardly begun, suddenly
drops his work and runs to the moving ball, to lend a
hand to the owner. His aid seems to be accepted
[13]
FABRE'S BOOK OF INSECTS
willingly. But the new-comer is not really a partner:
he is a robber. To make one's own ball needs hard work
and patience; to steal one ready-made, or to invite one-
self to a neighbour's dinner, is much easier. Some thiev-
ing Beetles go to work craftily, others use violence.
Sometimes a thief comes flying up, knocks over the
owner of the ball, and perches himself on top of it.
With his fore-legs crossed over his breast, ready to hit
out, he awaits events. If the owner raises himself to
seize his ball the robber gives him a blow that stretches
him on his back. Then the o\vner gets up and shakes
the ball till it begins rolling, and perhaps the thief falls
oif. A wrestling-match follows. The two Beetles
grapple with one another: their legs lock and unlock,
their joints intertwine, their horny armour clashes and
grates with the rasping sound of metal under a file. The
one who is successful climbs to the top of the ball, and
after two or three attempts to dislodge him the defeated
Scarab goes off to make himself a new pellet. I have
sometimes seen a third Beetle appear, and rob the robber.
But sometimes the thief bides his time and trusts to
cunning. He pretends to help the victim to roll the
food along, over sandy plains thick with thyme, over
cart-ruts and steep places, but he really does very little
of the work, preferring to sit on the ball and do nothing.
When a suitable place for a burrow is reached the right-
[14]
THE SACRED BEETLE
ful owner begins to dig with his sharp-edged forehead
and toothed legs, flinging armfuls of sand behind him,
while the thief clings to the ball, shamming dead. The
cave grows deeper and deeper, and the working Scarab
disappears from view. Whenever he comes to the sur-
face he glances at the ball, on which the other lies, de-
mure and motionless, inspiring confidence. But as the
absences of the owner become longer the thief seizes his
chance, and hurriedly makes off with the ball, which he
pushes behind him with the speed of a pickpocket afraid
of being caught. If the owner catches him, as some-
times happens, he quickly changes his position, and seems
to plead as an excuse that the pellet rolled down the
slope, and he was only trying to stop it I And the two
bring the ball back as though nothing had happened.
If the thief has managed to get safely away, however,
the owner can only resign himself to his loss, which he
does with admirable fortitude. He rubs his cheeks,
sniffs the air, flies off, and begins his work all over again.
I admire and envy his character.
At last his provisions are safely stored. His burrow
is a shallow hole about the size of a man's fist, dug in
soft earth or sand, with a short passage to the surface,
just wide enough to admit the ball. As soon as his food
is rolled into this burrow the Scarab shuts himself in
by stopping up the entrance with rubbish. The ball
[15]
FABRE'S BOOK OF INSECTS
fills almost the whole room: the banquet rises from floor
to ceiling. Only a narrow passage runs between it and
the walls, and here sit the banqueters, two at most, very
often only one. Here the Sacred Beetle feasts day and
night, for a week or a fortnight at a time, without
ceasing.
II
THE PEAR
As I have already said, the ancient Egyptians thought
that the egg of the Sacred Beetle was within the ball
that I have been describing. I have proved that it
is not so. One day I discovered the truth about the
Scarab's egg.
A young shepherd who helps me in his spare time
came to me one Sunday in June with a queer thing in
his hand. It was exactly like a tiny pear that had lost
all its fresh colour and had turned brown in rotting. It
was firm to the touch and very graceful in shape, though
the materials of which it was formed seemed none too
nicely chosen. The shepherd assured me there was an
c^^ inside it; for a similar pear, crushed by accident in
the digging, had contained a white egg the size of a grain
of wheat.
[16]
THE SACRED BEETLE
At daybreak the next morning the shepherd and I went
out to investigate the matter. We met among the brows-
ing sheep, on some slopes that had lately been cleared
of trees.
A Sacred Beetle's burrow is soon found : you can tell
it by the fresh little mound of earth above it. My com-
panion dug vigorously into the ground with my pocket
trowel, while I lay down, the better to see what was being
unearthed. A cave opened out, and there I saw, lying
in the moist earth, a splendid pear upon the ground. I
shall not soon forget my first sight of the mother Beetle's
wonderful work. My excitement could have been no
greater had I, in digging among the relics of ancient
Egypt, found the sacred insect carved in emerald.
We went on with our search, and found a second hole.
Here, by the side of the pear and fondly embracing it,
was the mother Beetle, engaged no doubt in giving it
the finishing touches before leaving the burrow for good.
There was no possible doubt that the pear was the nest
of the Scarab. In the course of the summer I found at
least a hundred such nests.
The pear, like the ball, is formed of refuse scraped
up in the fields, but the materials are less coarse, because
they are intended for the food of the grub. When it
comes out of the egg it is incapable of searching for its
[17]
FABRE'S BOOK OF INSECTS
own meals, so the mother arranges that it shall find itself
surrounded by the food that suits it best. It can begin
eating at once, without further trouble.
Hie egg is laid in the narrow end of the pear. Every
germ of life, whether of plant or animal, needs air: even
tlie shell of a bird's egg is riddled with an endless number
of pores. If the germ of the Scarab were in the thick
part of the pear it would be smothered, because there the
materials are very closely packed, and are covered with
a hard rind. So the mother Beetle prepares a nice airy
room with thin walls for her little grub to live in, during
its first moments. There is a certain amount of air even
in the very centre of the pear, but not enough for a deli-
cate baby-grub. By the time he has eaten his way to the
centre he is strong enough to manage with very little air.
There is, of course, a good reason for the hardness of
the shell that covers the big end of the pear. The
Scarab's burrow is extremely hot: sometimes the tem-
perature reaches boiling point. The provisions, even
though they have to last only three or four weeks, are
liable to dry up and become uneatable. When, instead
of the soft food of its first meal, the unhappy grub finds
nothing to eat but horrible crusty stuff as hard as a peb-
ble, it is bound to die of hunger. I have found numbers
of these victims of the August sun. The poor things
are baked in a sort of closed oven. To lessen this danger
[•8]
THE SACRED BEETLE
the mother Beetle compresses the outer layer of the pear
— or nest — with all the strength of her stout, flat fore-
arms, to turn it into a protecting rind like the shell of
a nut. This helps to ward off the heat. In the hot
summer months the housewife puts her bread into a
closed pan to keep it fresh. The insect does the same in
its own fashion : by dint of pressure it covers the family
bread with a pan.
I have watched the Sacred Beetle at work in her den,
so I know how she makes her pear-shaped nest.
With the building-materials she has collected she
shuts herself up underground so as to give her whole at-
tention to the business in hand. The materials may be
obtained in two ways. As a rule, under natural condi-
tions, she kneads a ball in the usual way and rolls it to
a favourable spot. As it rolls along it hardens a little
on the surface and gathers a slight crust of earth and
tiny grains of sand, which is useful later on. Now and
then, however, the Beetle finds a suitable place for her
burrow quite close to the spot where she collects her
building-materials, and in that case she simply bundles
armfuls of stuff into the hole. The result is most strik-
ing. One day I see a shapeless lump disappear into the
burrow. Next day, or the day after, I visit the Beetle's
workshop and find the artist in front of her work. The
[19]
FABRE'S BOOK OF INSECTS
formless mass of scrapings has become a pear, perfect
in outline and exquisitely finished.
The part that rests on the floor of the burrow is crusted
over with particles of sand, while the rest is polished
like pi ass. This shows that the Beetle has not rolled
the pear round and round, but has shaped it where it
lies. She has modelled it with little taps of her broad
feet, just as she models her ball in the daylight.
By making an artificial burrow for the mother Beetle
in my own workshop, with the help of a glass jar full
of earth, and a peep-hole through which I can observe
operations, I have been able to see the work in its vari-
ous stages.
The Beetle first makes a complete ball. Then she
starts the neck of the pear by making a ring round the
hall and applying pressure, till the ring becomes a groove.
In this way a blunt projection is pushed out at one side
of the ball. In the centre of this projection she employs
further pressure to form a sort of crater or hollow, with
a swollen rim; and gradually the hollow is made deeper
and the swollen rim thinner and thinner, till a sack is
formed. In this sack, which is polished and glazed in-
side, the egg is laid. The opening of the sack, or extreme
end of the pear, is then closed with a plug of stringy
fibres.
There is a reason for this rough plug — a most curious
[20]
THE SACRED BEETLE
exception, when nothing else has escaped the heavy blows
of the insect's leg. The end of the egg rests against it,
and, if the stopper were pressed down and driven in,
the infant grub might suffer. So the Beetle stops the
hole without ramming down the stopper.
Ill
THE GROWING-UP OF THE SCARAB
About a week or ten days after the laying of the egg,
the grub is hatched, and without delay begins to eat its
house. It is a grub of remarkable wisdom, for it always
starts its meal with the thickest part of the walls, and
so avoids making a hole through which it might fall out
of the pear altogether. It soon becomes fat; and indeed
it is an ungainly creature at best, with an enormous
hump on its back, and a skin so transparent that if you
hold it up to the light you can see its internal organs.
If the early Egyptian had chanced upon this plump
white grub he would never have suspected it to contain,
in an undeveloped state, the sober beauty of the Scarab !
When first it sheds its skin the insect that appears
is not a full-grown Scarab, though all the Scarab's
features can be recognised. There are few insects so
beautiful as this delicate creature with its wing-cases
Iving in front of it like a wide pleated scarf and its fore-
[21]
FABRE'S BOOK OF INSECTS
legs folded under its head. Half transparent and as
yellow as honey, it looks as though it were carved from
a block of amber. For four weeks it remains in this
state, and then it too casts its skin.
Its colouring now is red-and-white, — so many times
does the Sacred Beetle change its garments before it
finally appears black as ebony I As it grows blacker it
also grows harder, till it is covered with horny armour
and is a full-grown Beetle.
All this time he is underground, in the pear-shaped
nest. Great is his longing to burst the shell of his prison
and come into the sunshine. Whether he succeeds in
doing so depends on circumstances.
It is generally August when he is ready for release,
and August as a rule is the driest and hottest month of
the year. If therefore no rain falls to soften the earth,
the cell to be burst and the wall to be broken defy the
strength of the insect, which is helpless against all that
hardness. The soft material of the nest has become an
impassable rampart; it has turned into a sort of brick,
baked in the kiln of summer.
I have, of course, made experiments on insects that
are ready to be released. I lay the hard, dry shells in
a box where they remain dry; and sooner or later I hear
a sharp, grating sound inside each cell. It is the
prisoner scraping the wall with the rakes on his fore-
[22]
THE SACRED BEETLE
head and his fore-feet. Two or three days pass, and
no progress seems to have been made.
I try to help a couple of them by opening a loophole
with my knife; but these favoured ones make no more
progress than the others.
In less than a fortnight silence reigns in all the shells.
The prisoners, worn out with their efforts, have all died.
Then I take some other shells, as hard as the first,
wrap them in a wet rag, and put them in a corked flask.
"When the moisture has soaked through them I rid them
of the wrapper, but keep them in the flask. This time
the experiment is a complete success. Softened by the
wet the shells are burst by the prisoner, who props him-
self boldly on his legs, using his back as a lever, or else
scrapes away at one point till the walls crumble to pieces.
In every case the Beetle is released.
In natural conditions, when the shells remain under-
ground, the same thing occurs. When the soil is burnt
by the August sun it is impossible for the insect to wear
away his prison, which is hard as a brick. But when
a shower comes the shell recovers the softness of its early
days : the insect struggles with his legs and pushes with
his back, and so becomes free.
At first he shows no interest in food. What he wants
above all is the joy of the light. He sets himself in
the sun, and there, motionless, basks in the warmth.
[23]
FABRE'S BOOK OF INSECTS
Presently, however, he wishes to eat. With no one
to teach him, he sets to work, exactly like his elders, to
make himself a ball of food. He digs his burrow and
stores it with provisions. Without ever learning it, he
knows his trade to perfection.
[24]
CHAPTER III
THE CICADA
THE CICADA AND THE ANT
TO most of US the Cicada's song is unknown,
for he lives in the land of the olive-trees.
But every one who has read La Fontaine's
"Fables" has heard of the snub the Cicada received from
the Ant, though La Fontaine was not the first to tell
the tale.
The Cicada, says the story, did nothing but sing all
through the summer, while the Ants were busy storing
their provisions. When winter came he was hungry,
and hurried to his neighbour to borrow some food. He
met with a poor welcome.
"Why didn't you gather your food in the summer?"
asked the prudent Ant.
"I was busy singing all the summer,' said the Cicada.
"Singing, were you*?" answered the Ant unkindly.
"Well, then, now you may dance I" And she turned her
back on the beggar.
[25]
FABRE'S BOOK OF INSECTS
Now the insect in this fable could not possibly be
a Cicada. La Fontaine, it is plain, was thinking of the
Grasshopper and as a matter of fact the English trans-
lations usually substitute a Grasshopper for the Cicada.
For my village does not contain a peasant so ignorant
as to imagine the Cicada ever exists in winter. Every
tiller of the soil is familiar with the grub of this insect,
which he turns over with his spade whenever he banks
up the olive-trees at the approach of cold weather. A
thousand times he has seen the grub leave the ground
through a round hole of its own making, fasten itself to
a twig, split its own back, take off its skin, and turn into
a Cicada.
The fable is a slander. The Cicada is no beggar,
though it is true that he demands a good deal of atten-
tion from his neighbours. Every summer he comes and
settles in his hundreds outside my door, amid the
greenery of two tall plane-trees; and here, from sun-
rise to sunset, he tortures my head with the rasping of
his harsh music. This deafening concert, this incessant
rattling and drumming, makes all thought impossible.
It is true, too, thiat there are sometimes dealings
between the Cicada and the Ant; but they are exactly
the opposite of those described in the fable. The Cicada
is never dependent on others for his living. At no time
does he go crying famine at the doors of the Ant-hills.
[26]
THE CICADA
On the contrary, it is the Ant who, driven by hunger,
begs and entreats the singer. Entreats, did I say^ It
is not the right word. She brazenly robs him.
In July, when most of the insects in my sunny country
are parched with thirst, and vainly wander round the
withered flowers in search of refreshment, the Cicada
remains perfectly cheerful. With his rostrum — the deli-
cate sucker, sharp as a gimlet, that he carries on his chest
I — he broaches a cask in his inexhaustible cellar. Sitting,
always singing, on the branch of a shrub, lie bores through
the firm, smooth bark, which is swollen with sap. Driv-
ing his sucker through the bunghole, he drinks his fill.
If I watch him for a little while I may perhaps see
him in unexpected trouble. There are many thirsty
insects in the neighbourhood, who soon discover the sap
that oozes from the Cicada's well. They hasten up, at
first quietly and discreetly, to lick the fluid as it comes
out. I see Wasps, Flies, Earwigs, Rose-chafers, and
above all. Ants.
The smallest, in order to reach the well, slip under
the body of the Cicada, who good-naturedly raises him-
self on his legs to let them pass. The larger insects
snatch a sip, retreat, take a walk on a neighbouring
branch, and then return more eager and enterprising
than before. They now become violent brigands, deter-
mined to chase the Cicada away from his well.
[27]
FABRE'S BOOK OF INSECTS
The worst offenders are the Ants. I have seen them
nibbling at the ends of the Cicada's legs, tugging at the
tips of his wings, and climbing on his back. Once a
bold robber, before my very eyes, caught hold of a Ci-
cada's sucker and tried to pull it out.
At last, worried beyond all patience, the singer deserts
the well he has made. The Ant has now attained her
object: she is left in possession of the spring. This
dries up very soon, it is true; but, having drunk all the
sap that is there, she can wait for another drink till she
has a chance of stealing another well.
So you see that the actual facts are just the reverse
of those in the fable. The Ant is the hardened beggar :
the industrious worker is the Cicada.
II
THE cicada's burrow
I am in an excellent position to study the habits of the
Cicada, for I live in his company. When July comes he
takes possession of the enclosures right up to the threshold
of the house. I remain master indoors, but out of doors
he reigns supreme, and his reign is by no means a peace-
ful one.
The first Cicadae appear at midsummer. In the much-
trodden, sun-baked paths I see, level with the ground,
[28]
THE CICADA
round holes about the size of a man's thumb. Through
these holes the Cicada-grubs come up from the under-
ground to be transformed into full-grown Cicadse on the
surface. Their favourite places are the driest and
sunniest; for these grubs are provided with such powerful
tools that they can bore through baked earth or sandstone.
When I examine their deserted burrows I have to use
my pickaxe.
The first thing one notices is that the holes, which
measure nearly an inch across, have absolutely no rubbish
round them. There is no mound of earth thrown up
outside. Most of the digging insects, such as the Dor-
beetles for instance, make a mole-hill above their
burrows. The reason for this difference lies in their
manner of working. The Dorbeetle begins his work at
the mouth of the hole, so he can heap up on the surface the
material he digs out: but the Cicada-grub comes up
from below. The last thing he does is to make the door-
way, and he cannot heap rubbish on a threshold that does
not yet exist.
The Cicada's tunnel runs to a depth of fifteen or six-
teen inches. It is quite open the whole way. It ends
in a rather wider space, but is completely closed at the
bottom. What has become of the earth removed to make
this tunnel ? And why do not the walls crumble ? One
would expect that the grub, climbing up and down with
[29]
FABRE'S BOOK OF INSECTS
his clawed legs, would make landslips and block up
his own house.
Well, he behaves like a miner or a railway-engineer.
The miner holds up his galleries with pit-props; the
builder of railways strengthens his tunnel with a casing
of brickwork; the Cicada is as clever as either of them,
and covers the walls of his tunnel with cement. He
carries a store of sticky fluid hidden within him, with
which to make this plaster. His burrow is always built
above some tiny rootlet containing sap, and from this root
he renews his supply of fluid.
It is very important for him to be able to run up and
down his burrow at his ease, because, when the time comes
for him to find his way into the sunshine, he wants to
know what the weather is like outside. So he works away
for weeks, perhaps for months, to make a funnel with
good strong plastered walls, on which he can clamber.
At the top he leaves a layer as thick as one's finger, to
protect him from the outer air till the last moment. At
the least hint of fine weather he scrambles up, and,
through the thin lid at the top, inquires into the state of
the weather.
If he suspects a storm or rain on the surface — matter of
great importance to a delicate grub when he takes off his
skin I — he slips prudently back to the bottom of his snug
funnel. But if the weather seems warm he smashes his
[30]
THE CICADA
ceiling with a few strokes of his claws, and climbs to
the surface.
It is the fluid substance carried by the Cicada-grub in
his swollen body that enables him to get rid of the rubbish
in his burrow. As he digs he sprinkles the dusty earth
and turns it into paste. The walls then become soft and
yielding. The mud squeezes into the chinks of the rough
soil, and the grub compresses it with his fat body. This
is why, when he appears at the top, he is always covered
with wet stains.
For some time after the Cicada-grub's first appearance
above-ground he wanders about the neighbourhood,
looking for a suitable spot in which to cast off his skin — a
tiny bush, a tuft of thyme, a blade of grass, or the twig of
a shrub. When he finds it he climbs up, and clings to
it firmly with the claws of his fore-feet. His fore-legs
stiffen into an immovable grip.
Then his outer skin begins to split along the middle
of the back, showing the pale-green Cicada within.
Presently the head is free ; then the sucker and front legs
appear, and finally the hind-legs and the rumpled wings.
The whole insect is free now, except the extreme tip of
his body.
He next performs a wonderful gymnastic feat. High
in the air as he is, fixed to his old skin at one point
[31]
FABRE'S BOOK OF INSECTS
only, he turns himself over till his head is hanging
downwards. His crumpled wings straighten out, un-
furl, and spread themselves. Then with an almost in-
visible movement he draws himself up again by sheer
strength, and hooks his fore-legs on to his empty skin.
This movement has released the tip of his body from its
sheath. The whole operation has taken about half an
hour.
For a time the freed Cicada does not feel very strong.
He must bathe in air and sunshine before strength and
colour come to his frail body. Hanging to his cast skin
by his fore-claws only, he sways at the least breath of
air, still feeble and still green. But at last the brown
tinge appears, and is soon general. Supposing him to
have taken possession of the twig at nine o'clock in the
morning, the Cicada flies away at half-past twelve, leav-
ing his cast skin behind him. Sometimes it hangs from
the twigs for months.
Ill
THE cicada's music
The Cicada, it appears, loves singing for its own
sake. Not content with carrying an instrument called
the cymbal in a cavity behind his wings, he increases
its power by means of sounding-boards under his chest.
[32]
THE CICADA
Indeed, there is one kind of Cicada who sacrifices a great
deal in order to give full play to his musical tastes. He
carries such an enormous sounding-board that there is
hardly any room left for his vital organs, which are
squeezed into a tiny corner. Assuredly one must be
passionately devoted to music thus to clear away one's
internal organs in order to make room for a musical box I
Unfortunately the song he loves so much is extremely
unattractive to others. Nor have I yet discovered its
object. It is usually suggested that he is calling his
mate ; but the facts appear to contradict this idea.
For fifteen years the Common Cicada has thrust his
society upon me. Every summer for two months I
have these insects before my eyes, and their song in my
ears. I see them ranged in rows on the smooth bark of
the plane-trees, the maker of music and his mate sitting
side by side. With their suckers driven into the tree
they drink, motionless. As the sun turns they also turn
round the branch with slow, sidelong steps, to find the
hottest spot. Whether drinking or moving they never
cease singing.
It seems unlikely, therefore, that they are calling
their mates. You do not spend months on end calling
to some one who is at your elbow.
Indeed, I am inclined to think that the Cicada him-
[33]
FABRE'S BOOK OF INSECTS
self cannot even hear the song he sings with so much
apparent delight. This might account for the relentless
\va} in which he forces his music upon others.
He has very clear sight. His five eyes tell him what
is happening to right and to left and above his head;
and the moment he sees any one coming he is silent and
flies away. Yet no noise disturbs him. Place yourself
behind him, and then talk, whistle, clap your hands,
and knock two stones together. For much less than this
a bird, though he would not see you, would fly away
terrified. The imperturbable Cicada gones on rattling
as though nothing were there.
On one occasion I borrowed the local artillery, that
is to say the guns that are fired on feast-days in the vil-
lage. There were two of them, and they were crammed
with powder as though for the most important rejoicings.
They were placed at the foot of the plane-trees in front
of ni}- door. We were careful to leave the windows
open, to prevent the panes from breaking. The Cicadse
m the branches overhead could not see what was
hap[)ening.
Six of us waited below, eager to hear what would be
the effect on the orchestra above.
Banff! The gun went off with a noise like a thunder-
clap.
Quite unconcerned, the Cicadas continued to sing.
[34]
THE CICADA
Not one appeared in the least disturbed. There was
no change whatever in the quality or the quantity of
the sound. The second gun had no more effect than the
first.
I think, after this experiment, we must admit that the
Cicada is hard of hearing, and like a very deaf man, ii,
quite unconscious that he is making a noise.
IV
THE CICADA S EGGS
The Common Cicada likes to lay her eggs on smull
dry branches. She chooses, as far as possible, tiny
stalks, which may be of any size between that of a straw
and a lead-pencil. The sprig is never lying on the
ground, is usually nearly upright in position, and is al-
most always dead.
Having found a twig to suit her, she makes a row of
pricks with the sharp instrument on her chest — such
pricks as might be made with a pin if it were driven
downwards on a slant, so as to tear the fibres and force
them slightly upwards. If she is undisturbed she will
make thirty or forty of these pricks on the same twig.
In the tiny cells formed by these pricks she lays her
eggs. The cells are narrow passages, each one slanting
down towards the one below it. I generally find about
[35]
FABRE'S BOOK OF INSECTS
ten eggs in each cell, so it is plain that the Cicada lays
between three and four hundred eggs altogether.
This is a fine family for one insect. The numbers
jK)int to some special danger that threatens the Cicada,
and HKikcs it necessary to produce a great quantity of
grubs lest some should be destroyed. After many obser-
vations I have discovered what this danger is. It is
an extremely tiny Gnat, compared with which the
Cicada is a monster.
This Gnat, like the Cicada, carries a boring-tool. It
is planted beneath her body, near the middle, and sticks
out at right angles. As fast as the Cicada lays her eggs
the Gnat tries to destroy them. It is a real scourge to
the Cicada family. It is amazing to watch her calm and
brazen audacity in the presence of the giant who could
crush her by simply stepping on her. I have seen as
many as three preparing to despoil one unhappy Cicada
at the same time, standing close behind one another.
The Cicada has just stocked a cell with eggs, and is
climbing a little higher to make another cell. One of
the brigands runs to the spot she has just left; and here,
almost under the claws of the monster, as calmly and
fearlessly as though she were at home, the Gnat bores
a second hole above the Cicada's eggs, and places among
them an egg of her own. By the time the Cicada flies
away most of her cells have, in this way, received a
[36]
THE CICADA
stranger's egg, which will be the ruin of hers. A small
quick-hatching grub, one only to each cell, handsomely
fed on a dozen raw eggs, will take the place of the
Cicada's family.
This deplorable mother has learnt nothing from
centuries of experience. Her large and excellent eyes
cannot fail to see the terrible felons fluttering round her.
She must know they are at her heels, and yet she remains
unmoved, and lets herself be victimised. She could
easily crush the wicked atoms, but she is incapable of
altering her instincts, even to save her family from
destruction.
Through my magnifying-glass I have seen the hatch-
ing of the Cicada's eggs. When the grub first appears
it has a marked likeness to an extremely small fish, with
large black eyes, and a curious sort of mock fin under
its body, formed of the two fore-legs joined together.
This fin has some power of movement, and helps the
grub to work its way out of the shell, and also — a much
more difBcult matter — out of the fibrous stem in which
it is imprisoned.
As soon as this fish-like object has made its way out
of the cell it sheds its skin. But the cast skin forms
itself into a thread, by which the grub remains fastened
to the twig or stem. Here, before dropping to the
ground, it treats itself to a sun-bath, kicking about and
[37]
FABRE'S BOOK OF INSECTS
trying its strength, or swinging lazily at the end of its
rope.
Its antennae now are free, and wave about; its legs
work their joints; those in front open and shut their
chiws. I know hardly any more curious sight than this
tiny acrobat hanging by the tip of its body, swinging at
the least breath of wind, and making ready in the air
for its somersault into the world.
Sooner or later, without losing much time, it drops
to the ground. The little creature, no bigger than a
Flea, has saved its tender body from the rough earth
by swinging on its cord. It has hardened itself in the
air, that luxurious eiderdown. It now plunges into the
stern realities of life.
I see a thousand dangers ahead of it. The merest
breath of wind could blow it on to the hard rock, or into
the stagnant water in some deep cart-rut, or on the
sand where nothing grows, or else on a clay soil, too
tough for it to dig in.
The feeble creature needs shelter at once, and must
look for an underground refuge. The days are growing
cold, and delays are fatal to it. It must wander about
in search of soft soil, and no doubt many die before
they find it.
When at last it discovers the right spot it attacks the
earth with the hooks on its fore-feet. Through the mag-
[38]
THE CICADA
nifying-glass I watch it wielding its pickaxes, and raking
an atom of earth to the surface. In a few minutes a
well has been scooped out. The little creature goes
down into it, buries itself, and is henceforth invisible.
The underground life of the undeveloped Cicada
remains a secret. But we know how long it remains
in the earth before it comes to the surface and becomes
a full-grown Cicada. For four years it lives below the
soil. Then for about five weeks it sings in the sunshine.
Four years of hard work in the darkness, and a month
of delight in the sun — such is the Cicada's life. We
must not blame him for the noisy triumph of his song.
For four years he has dug the earth with his feet, and
then suddenly he is dressed in exquisite raiment, pro-
vided with wings that rival the bird's, and bathed in
heat and light I What cymbals can be loud enough
to celebrate his happiness, so hardly earned, and so
very, very short?
[39]
CHAPTER IV
THE PRAYING MANTIS
I
HER HUNTING
THERE is an insect of the south that is quite as
interesting as the Cicada, but much less famous,
because it makes no noise. Had it been pro-
vided with cymbals, its renown would have been greater
than the celebrated musician's, for it is most unusual
both in shape and habits.
A long time ago, in the days of ancient Greece, this
insect was named Mantis, or the Prophet. The peasant
saw her on the sun-scorched grass, standing half-erect
in a very imposing and majestic manner, with her broad
green gossamer wings trailing like long veils, and her
fore-legs, like arms, raised to the sky as though in prayer.
To the peasant's ignorance the insect seemed like a
priestess or a nun, and so she came to be called the
Praying Mantis.
There was never a greater mistake! Those pious
airs are a fraud; those arms raised in prayer are really
the most horrible weapons, which slay whatever passes
[40]
THE PRAYING MANTIS
within reach. The Mantis is fierce as a tigress, cruel as
an ogress. She feeds only on living creatures.
There is nothing in her appearance to inspire dread.
She is not without a certain beauty, with her slender,
graceful figure, her pale-green colouring, and her long
gauze wings. Having a flexible neck, she can move her
head freely in all directions. She is the only insect that
can direct her gaze wherever she will. She almost has
a face.
Great is the contrast between this peaceful-looking
body and the murderous machinery of the fore-legs.
The haunch is very long and powerful, while the thigh
is even longer, and carries on its lower surface two rows
of sharp spikes or teeth. Behind these teeth are three
spurs. In short, the thigh is a saw with two blades,
between which the leg lies when folded back.
This leg itself is also a double-edged saw, provided
with a greater number of teeth than the thigh. It ends
in a strong hook with a point as sharp as a needle, and
a double blade like a curved pruning-knife. I have
many painful memories of this hook. Many a time,
when Mantis-hunting, I have been clawed by the insect
and forced to ask somebody else to release me. No in-
sect in this part of the world is so troublesome to handle.
The Mantis claws you with her pruning-hooks, pricks you
with her spikes, seizes you in her vice, and makes self-
[41]
FABRE'S BOOK OF INSECTS
defence impossible if you wish to keep your captive alive.
When at rest, the trap is folded back against the
chest and looks quite harmless. There you have the
insect praying. But if a victim passes by, the appear-
ance of prayer is quickly dropped. The three long
divisions of the trap are suddenly unfolded, and the
prey is caught with the sharp hook at the end of them,
and drawn back between the two saws. Then the vice
closes, and all is over. Locusts, Grasshoppers, and
even stronger insects are helpless against the four rows
of teeth.
It is impossible to make a complete study of the habits
of the Mantis in the open fields, so I am obliged to take
her indoors. She can live quite happily in a pan filled
with sand and covered with a gauze dish-cover, if only
she be supplied with plently of fresh food. In order to
find out what can be done by the strength and daring
of the Mantis, I provide her not only with Locusts and
Grasshoppers, but also with the largest Spiders of the
neiglibourhood. This is what I see.
A grey Locust, heedless of danger, walks towards the
Mantis. The latter gives a convulsive shiver, and sud-
denly, in the most surprising way, strikes an attitude
that fills the Locust with terror, and is quite enough to
startle any one. You see before you unexpectedly a
sort of bogy-man or Jack-in-the-box. The wing-covers
[42]
V'
%
/
^
V
''V^
"X
V ^
THE PRAYING MANTIS
open; the wings spread to their full extent and stand
erect like sails, towering over the insect's back; the tip
of the body curls up like a crook, rising and falling with
short jerks, and making a sound like the puffing of a
startled Adder. Planted defiantly on its four hind-legs,
the Mantis holds the front part of its body almost up-
right. The murderous legs open wide, and show a pat-
tern of black-and-white spots beneath them.
In this strange attitude the Mantis stands motionless,
with eyes fixed on her prey. If the Locust moves, the
Mantis turns her head. The object of this performance
is plain. It is intended to strike terror into the heart
of the victim, to paralyse it with fright before attacking
it. The Mantis is pretending to be a ghost!
The plan is quite successful. The Locust sees a
spectre before him, and gazes at it without moving. He
to whom leaping is so easy makes no attempt at escape.
He stays stupidly where he is, or even draws nearer with
a leisurely step.
As soon as he is within reach of the Mantis she strikes
with her claws; her double saws close and clutch; the
poor wretch protests in vain; the cruel ogress begins her
meal.
The pretty Crab Spider stabs her victim in the neck,
in order to poison it and make it helpless. In the same
way the Mantis attacks the Locust first at the back of the
[43]
FABRE'S BOOK OF INSECTS
neck, to destroy its power of movement. This enables
her to kill and eat an insect as big as herself, or even
bigger. It is amazing that the greedy creature can con-
tain so much food.
The various Digger-wasps receive visits from her
pretty frequently. Posted near the burrows on a bram-
ble, she waits for chance to bring near her a double prize,
the Hunting-wasp and the prey she is bringing home.
For a long time she waits in vain ; for the Wasp is sus-
picious and on her guard: still, now and then a rash
one is caught. With a sudden rustle of wings the
Mantis terrifies the new-comer, who hesitates for a mo-
ment in her fright. Then, with the sharpness of a
spring, the Wasp is fixed as in a trap between the blades
of the double saw — the toothed fore-arm and toothed
upper-arm of the Mantis. The victim is then gnawed
in small mouthfuls.
I once saw a Bee-eating Wasp, while carrying a Bee
to her storehouse, attacked and caught by a Mantis.
The Wasp was in the act of eating the honey she had
found in the Bee's crop. The double saw of the Mantis
closed suddenly on the feasting Wasp; but neither terror
nor torture could persuade that greedy creature to leave
off eating. Even while she was herself being actually
devoured she continued to lick the honey from her Bee !
I regret to say that the meals of this savage ogress
[44]
THE PRAYING MANTIS
are not confined to other kinds of insects. For all her
sanctimonious airs she is a cannibal. She will eat her
sister as calmly as though she were a Grasshopper; and
those around her will make no protest, being quite ready
to do the same on the first opportunity. Indeed, she even
makes a habit of devouring her mate, whom she seizes
by the neck and then swallows by little mouthfuls, leav-
ing only the wings.
She is worse than the Wolf; for it is said that even
Wolves never eat each other.
II
HER NEST
After all, however, the Mantis has her good points,
like most people. She makes a most marvellous nest.
This nest is to be found more or less everywhere in
sunny places: on stones, wood, vine-stocks, twigs,
or dry grass, and even on such things as bits of brick,
strips of linen, or the shrivelled leather of an old boot.
Any support will serve, as long as there is an uneven
surface to form a solid foundation.
In size the nest is between one and two inches long,
and less than an inch wide; and its colour is as golden
as a grain of wheat. It is made of a frothy substance,
which has become solid and hard, and it smells like silk
[45]
FABRE'S BOOK OF INSECTS
when it is burnt. The shape of it varies according to
the support on which it is based, but in all cases the upper
surface is convex. One can distinguish three bands,
or zones, of which the middle one is made of little plates
or scales, arranged in pairs and over-lapping like the
tiles of a roof. The edges of these plates are free,
forming two rows of slits or little doorways, through
which the young Mantis escapes at the moment of hatch-
ing. In every other part the wall of the nest is impene-
trable.
The eggs are arranged in layers, with the ends con-
taining the heads pointed towards the doorways. Of
these doorways, as I have just said, there are two rows.
One half of the grubs will go out through the right door,
and the other half through the left.
It is a remarkable fact that the mother Mantis builds
this cleverly-made nest while she is actually laying her
eggs. From her body she produces a sticky substance,
rather like the Caterpillar's silk-fluid; and this material
she mixes with the air and whips into froth. She beats
it into foam with two ladles that she has at the tip of her
ImxIv, just as we beat white of egg with a fork. The
foam is greyish-white, almost like soapsuds, and when
It first appears it is sticky; but two minutes afterwards
it has solidified.
In this sea of foam the Mantis deposits her eggs. As
[46]
THE PRAYING MANTIS
each layer of eggs is laid, it is covered with froth, which
quickly becomes solid.
In a new nest the belt of exit-doors is coated with
a material that seems different from the rest — a layer
of fine porous matter, of a pure, dull, almost chalky
white, which contrasts with the dirty white of the remain-
der of the nest. It is like the mixture that confectioners
make of whipped white of egg, sugar, and starch, with
which to ornament their cakes. This snowy covering
is very easily crumbled and removed. When it is gone
the exit-belt is clearly visible, with its two rows of
plates. The wind and rain sooner or later remove it in
strips or flakes, and therefore the old nests show no traces
of it.
But these two materials, though they appear different,
are really only two forms of the same matter. The
Mantis with her ladles sweeps the surface of the foam,
skimming the top of the froth, and collecting it into a
band along the back of the nest. The ribbon that looks
like sugar-icing is merely the thinnest and lightest por-
tion of the sticky spray, which appears whiter than the
nest because its bubbles are more delicate, and reflect
more light.
It is truly a wonderful piece of machinery that can,
so methodically and swiftly, produce the horny central
substance on which the first eggs are laid, the eggs them-
[47]
FABRE'S BOOK OF INSECTS
selves, the protecting froth, the soft sugar-like covering
of the doorways, and at the same time can build over-
lapping plates, and the narrow passages leading t!0 them I
Yet the Mantis, while she is doing all this, hangs mo-
tionless on the foundation of the nest. She gives not
a glance at the building that is rising behind her. Her
legs act no part in the affair. The machinery works by
itself.
As soon as she has done her work the mother with-
draws. I expected to see her return and show some
tender feeling for the cradle of her family, but it
evidently has no further interest for her.
The Mantis, I fear, has no heart. She eats her hus-
band, and deserts her children.
Ill
THE HATCHING OF HER EGGS
The eggs of the Mantis usually hatch in bright sun-
shine, at about ten o'clock on a mid-June morning.
As I have already told you, there is only one part of
the nest from which the grub can find an outlet, namely
the band of scales round the middle. From under each
of these scales one sees slowly appearing a blunt, trans-
parent lump, followed by two large black specks, which
are the creature's eyes. The baby grub slips gently
[48]
THE PRAYING MANTIS
under the thin plate and half releases itself. It is
reddish yellow, and has a thick, swollen head. Under
its outer skin it is quite easy to distinguish the large
black eyes, the mouth flattened against the chest, the
legs plastered to the body from front to back. With
the exception of these legs the whole thing reminds one
somewhat of the first state of the Cicada on leaving the
egg.
Like the Cicada, the young Mantis finds it necessary
to wear an overall when it is coming into the world,
for the sake of convenience and safety. It has to emerge
from the depths of the nest through narrow, winding
ways, in which full-spread slender limbs could not find
enough room. The tall stilts, the murderous harpoons,
the delicate antennae, would hinder its passage, and
indeed make it impossible. The creature therefore
appears in swaddling-clothes, and has the shape of a
boat.
When the grub peeps out under the thin scales of its
nest its head becomes bigger and bigger, till it looks like
a throbbing blister. The little creature alternately
pushes forward and draws back, in its efforts to free it-
self, and at each movement the head grows larger. At
last the outer skin bursts at the upper part of the chest,
and the grub wriggles and tugs and bends about, deter-
mined to throw off its overall. Finallv the legs and the
[49]
FABRE'S BOOK OF INSECTS
lon^ antenn:t are freed, and a few shakes complete the
operation.
It is a striking sight to see a hundred young Mantes
coming from the nest at once. Hardly does one tiny
creature show its black eyes under a scale before a swarm
of others appears. It is as though a signal passed from
one to the other, so swiftly does the hatching spread.
Almost in a moment the middle zone of the nest is
covered with grubs, who run about feverishly, stripping
themselves of their torn garments. Then they drop off,
or clamber into the nearest foliage. A few days later
a fresh swarm appears, and so on till all the eggs are
hatched.
But alas I the poor grubs are hatched mto a world
of dangers. I have seen them hatching many times, both
out of doors in my enclosure, and in the seclusion of a
greenhouse, where I hoped I should be better able to
protect them. Twenty times at least I have watched
the scene, and every time the slaughter of the grubs
has been terrible. The Mantis lays many eggs,
but she will never lay enough to cope with the hungry
murderers who lie in wait until the grubs appear.
The Ants, above all, are their enemies. Every day
I find them visiting my nests. It is in vain for me to
interfere; they always get the better of me. They
seldom succeed in entering the nest; its hard walls form
[50]
THE PRAYING MANTIS
too strong a fortress. But they wait outside for their
prey.
The moment that the young grubs appear they are
grabed by the Ants, pulled out of their sheaths, and cut
in pieces. You see piteous struggles between the little
creatures who can only protest with wild wrigglings and
the ferocious brigands who are carrying them off. In
a moment the massacre is over; all that is left of the
flourishing family is a few scattered survivors who have
escaped by accident.
It is curious that the Mantis, the scourge of the insect
race, should be herself so often devoured at this early
stage of her life, by one of the least of that race, the
Ant. The ogress sees her family eaten by the dwarf.
But this does not continue long. So soon as she has
become firm and strong from contact with the air the
Mantis can hold her own. She trots about briskly among
the Ants, who fall back as she passes, no longer daring to
tackle her : with her fore-legs brought close to her chest,
like arms ready for self-defence, she already strikes awe
into them by her proud bearing.
But the Mantis has another enemy who is less easily
dismayed. The little Grey Lizard, the lover of sunny
walls, pays small heed to threatening attitudes. With
the tip of his slender tongue he picks up, one by one,
the few stra-'^ insects that have escaped the Ant. They
[51]
FABRE'S BOOK OF INSECTS
make but a small mouthful, but to judge from the
Lizard's expression they taste very good. Every time
he gulps down one of the little creatures he half-closes
his eyelids, a sign of profound satisfaction.
Moreover, even before the hatching the eggs are in
danger. There is a tiny insect called the Chalcis, who
carries a probe sharp enough to penetrate the nest of
solidified foam. So the brood of the Mantis shares the
fate of the Cicada's. The eggs of a stranger are laid
in the nest, and are hatched before those of the rightful
owner. The owner's eggs are then eaten by the in-
vaders. The Mantis lays, perhaps, a thousand eggs.
Possibly only one couple of these escapes destruction.
The Mantis eats the Locust : the Ant eats the Mantis :
the Wryneck eats the Ant. And in the autumn, when
the Wryneck has grown fat from eating many Ants, I
eat the Wryneck.
It may well be that the Mantis, the Locust, the Ant,
and even lesser creatures contribute to the strength of the
human brain. In strange and unseen ways they have
all supplied a drop of oil to feed the lamp of thought.
Their energies, slowly developed, stored up, and handed
on to us, pass into our veins and sustain our weakness.
We live by their death. The world is an endless circle.
Everything finishes so that everything may begin again;
everything dies so that everything may live.
[52]
THE PRAYING MANTIS
In many ages the Mantis has been regarded with super-
stitious awe. In Provence its nest is held to be the best
remedy for chilblains. You cut the thing in two, squeeze
it, and rub the afflicted part with the juice that streams
out of it. The peasants declare that it works like a
charm. I have never felt any relief from it myself.
Further, it is highly praised as a wonderful cure for
toothache. As long as you have it on you, you need
never fear that trouble. Our housewives gather it under
a favourable moon ; they keep it carefully in the corner of
a cupboard, or sew it into their pocket. The neighbours
borrow it when tortured by a tooth. They call it a tigno.
"Lend me your tigno \ I am in agony," says the sufferer
with the swollen face.
The other hastens to unstitch and hand over the
precious thing.
"Don't lose it, whatever you do," she says earnestly
to her friend. "It's the only one I have, and this isn't
the right time of moon."
This simplicity of our peasants is surpassed by an
English physician and man of science who lived in the
sixteenth century. He tells us that, in those days, if a
child lost his way in the country, he would ask the
Mantis to put him on his road. 'The Mantis," adds
the author, "will stretch out one of her feet and shew
him the right way and seldome or never misse."
[53]
CHAPTER V
THE GLOW-WORM
HIS SURGICAL INSTRUMENT
FEW insects enjoy more fame than the Glow-
worm, the curious little animal who celebrates
the joy of life by lighting a lantern at its tail-
end. We all know it, at least by name, even if we have
not seen it roaming through the grass, like a spark fallen
from the full moon. The Greeks of old called it the
Bright-tailed, and modern science gives it the name
Lampyris.
As a matter of fact the Lampyris is not a worm at all,
not even in general appearance. He has six short legs,
which he well knows how to use, for he is a real gad-
about. The male, when he is full-grown has wing-
cases, like the true Beetle that he is. The female is an
unattractive creature who knows nothing of the delights
of flying and all her life remains in the larva, or in-
complete form. Even at this stage the word "worm"
is out of place. We French use the phrase "naked as
[54]
THE GLOW-WORM
a worm" to express the lack of any kind of protection.
Now the Lampyris is clothed, that is to say he wears an
outer skin that serves as a defence; and he is, moreover,
rather richly coloured. He is dark brown, with pale
pink on the chest; and each segment, or division, of his
body is ornamented at the edge with two spots of fairly
bright red. A costume like this was never worn by a
worm I
Nevertheless we will continue to call him the Glow-
worm, since it is bv that name that he is best known to
the world.
The two most interesting peculiarities about the
Glow-worm are, first, the way he secures his food, and
secondly, the lantern at his tail.
A famous Frenchman, a master of the science of food,
once said :
"Show me what you eat, and I will tell you what you
are."
A similar question should be addressed to every insect
whose habits we propose to study; for the information
supplied by food is the chief of all the documents of
animal life. Well, in spite of his innocent appearance,
the Glow-worm is an eater of flesh, a hunter of game;
and he carries on his hunting with rare villainy. His
regular prey is the Snail. This fact has long been
known; but what is not so well known is his curious
[55]
FABRE'S BOOK OF INSECTS
method of attack, of which I have seen no other example
anywhere.
Before he begins to feed on his victim he gives it an
ana'sthetic — he makes it unconscious, as a person is
made unconscious with chloroform before a surgical
operation. His food, as a rule, is a certain small Snail
hardly the size of a cherry, which collects in clusters
during the hot weather, on the stiff stubble and other
dry stalks by the roadside, and there remains motion-
less, in profound meditation, throughout the scorching
summer days. In some such place as this I have often
seen the Glow-worm feasting on his unconscious prey,
which he had just paralysed on its shaky support.
But he frequents other places too. At the edge of
cool, damp ditches, where the vegetation is varied,
many Snails are to be found; and in such spots as these
the Glow-worm can kill his victim on the ground. I
can reproduce these conditions at home, and can there
follow the operator's performance down to the smallest
detail.
I will try to describe the strange sight. I place a
little grass in a wide glass jar. In this I install a few
Glow-worms and a supply of Snails of a suitable size,
neither too large nor too small. One must be patient
and wait, and above all keep a careful watch, for the
events take place unexpectedly and do not last long.
[56]
THE GLOW-WORM
For a moment the Glow-worm examines his prey,
which, according to its habit, is completely hidden in the
shell, except for the edge of the "mantle," which
projects slightly. Then the hunter draws his weapon.
It is a very simple weapon, but it cannot be seen without
a magnifying-glass. It consists of two mandibles, bent
back into a hook, very sharp and as thin as a hair.
Through the microscope one can see a slender groove
running down the hook. And that is all.
The insect repeatedly taps the Snail's mantle with its
instrument. It all happens with such gentleness as to
suggest kisses rather than bites. As children, teasing
one another, we used to talk of "tweaks" to express a
slight squeeze of the finger-tips, something more like
tickling than a serious pinch. Let us use that word.
In conversation with animals, language loses nothing by
remaining simple. The Glow-worm gives tweaks to
the Snail.
He doles them out methodically, without hurrying,
and takes a brief rest after each of them, as though to
find out what effect has been produced. The number
of tweaks is not great : half a dozen at most, which are
enough to make the Snail motionless, and to rob him of all
feeling. That other pinches are administered later, at
the time of eating, seems very likely, but I cannot say
anything for certain on that subject. The first few,
[57]
FABRE'S BOOK OF INSECTS
however — there are never many — are enough to prevent
the Snail from feeling anything, thanks to the prompti-
tude of the Glow-worm, who, at lightning speed, darts
some kind of poison into his victim by means of his
grooved hooks.
lliere is no doubt at all that the Snail is made in-
sensible to pain. If, when the Glow-worm has dealt
some four or five of his twitches, I take away the victim
and prick it with a fine needle, there is not a quiver in
the wounded flesh, there is not the smallest sign of life.
Moreover, I occasionally chance to see Snails attacked
by the Lampyris while they are creeping along the
ground, the foot slowly crawling, the tentacles swollen to
their full extent. A few disordered movements betray
a brief excitement on the part of the Snail, and then
everything ceases : the foot no longer crawls, the front-
part loses its graceful curve, the tentacles become limp
and give way under their own weight, dangling feebly
like a broken stick. The Snail, to all appearance, is
dead.
He is not, however, really dead. I can bring him to
life again. When he has been for two or three days in
a condition that is neither life nor death I give him a
shower-bath. In about a couple of days my prisoner,
so lately injured by the Glow-worm's treachery, is re-
stored to his usual state. He revives, he recovers move-
[58]
THE GLOW-WORM
ment and sensibility. He is affected by the touch of a
needle; he shifts his place, crawls, puts out his tentacles,
as though nothing unusual had occurred. The general
torpor, a sort of deep drunkenness, has vanished out-
right. The dead returns to life.
Human science did not invent the art of making a
person insensible to pain, which is one of the triumphs
of surgery. Far back in the centuries the Glow-worm,
and apparently others too, was practising it. The
surgeon makes us breathe the fumes of ether or chloro-
form: the insect darts forth from his fangs very tiny
doses of a special poison.
When we consider the harmless and peaceful nature
of the Snail it seems curious that the Glow-worm should
require this remarkable talent. But I think I know the
reason.
When the Snail is on the ground, creeping, or even
shrunk into his shell, the attack never presents any
difficulty. The shell possesses no lid and leaves the
hermit's fore-part to a great extent exposed. But it
very often happens that he is in a raised position, cling-
ing to the tip of a grass-stalk, or perhaps to the smooth
surface of a stone. This support to which he fastens
himself serves very well as a protection; it acts as a lid,
supposing that the shell fits closely on the stone or stalk.
But if the least bit of the Snail be left uncovered the
[59]
FABRE'S BOOK OF INSECTS
slender hooks of the Glow-worm can find their way in
through the gap, and in a moment the victim is made un-
conscious, and can be eaten in comfort.
Now, a Snail perched on top of a stalk is very easily
upset. The slightest struggle, the most feeble wriggle
on his part, would dislodge him; he would fall to the
ground, and the Glow-worm would be left without food.
It is necessary for the Snail to be made instantly un-
conscious of pain, or he would escape; and it must be
done with a touch so delicate that it does not shake him
from his stalk. And that, I think, is why the Glow-
worm possesses his strange surgical instrument.
II
HIS ROSETTE
The Glow-worm not only makes his victim insensible
while he is poised on the side of a dry grass-stalk, but
he eats him in the same dangerous posit'ion. And his
preparations for his meal are by no means simple.
What is his manner of consuming it? Does he
really eat, that is to say, does he divide his food into
pieces, does he carve it into minute particles, which are
afterwards ground by a chewing-apparatus? I think
not. I never see a trace of solid nourishment on my
[60]
THE GLOW-WORM
captives' mouths. The Glow-worm does not eat in the
strict sense of the word; he merely drinks. He feeds
on a thin gruel, into which he transforms his prey. Like
the flesh-eating grub of the Fly, he can digest his food
before he swallows it; he turns his prey into liquid
before feeding on it.
This is how things happen. A Snail has been made
insensible by a Glow-worm, who is nearly always alone,
even when the prize is a large one like the Common Snail.
Soon a number of guests hasten up^ — two, three, or more
— and, without any quarrel with the real owner, all alike
fall to. A couple of days later, if I turn the shell so
that the opening is downwards, the contents flow out like
soup from a saucepan. By the time the meal is finished
only insignificant remains are left.
The matter is obvious. By repeated tiny bites, similar
to the tweaks which we saw administered at the begin-
ning, the flesh of the Snail is converted into a gruel on
which the various guests nourish themselves each in his
own way, each working at the broth by means of some
special pepsine (or digestive fluid), and each taking his
own m.outhfuls of it. The use of this method shows that
the Glow-worm's mouth must be very feebly armed, apart
from the two fangs which sting the patient and inject
the poison. No doubt these fangs at the same time in-
[61]
FABRE'S BOOK OF INSECTS
ject some other substance which turns the solid flesh into
liquid, in such a thorough way that every morsel is turned
to account.
And this is done with exquisite delicacy, though some-
times in a position that is anything but steady. The
Snails imprisoned in my apparatus sometimes crawl up
to the top, which is closed with a glass pane. To this
pane they fix themseves with a speck of the sticky sub-
stance they carry with them; but, as they are miserly in
their use of this substance, the merest shake is enough
to loosen the shell and send it to the bottom of the jar.
Now it is not unusual for the Glow-worm to hoist
himself to the top, with the help of a certain climbing-
organ that makes up for the weakness of his legs. He
selects his prey, makes a careful inspection of it to find
a slit, nibbles it a little, makes it insensible, and then,
without delay, proceeds to prepare the gruel which he will
go on eating for days on end.
When he has finished his meal the shell is found to
be absolutely empty. And yet this shell, which was
fixed to the glass only by the slight smear of stickiness,
has not come loose, nor even shifted its position in the
smallest degree. Without any protest from the hermit
who has been gradually converted into broth, it has been
drained dry on the very spot at which the first attack was
made. These small details show us how promptly the
[62]
THE GLOW-WORM
anaesthetic bite takes effect, and how very skilfully the
Glow-worm treats his Snail.
To do all this, poised high in air on a sheet of glass
or a grass-stem, the Glow-worm must have some special
limb or organ to keep him from slipping. It is plain
that his short clumsy legs are not enough.
Through the magnifying-glass we can see that he does
indeed possess a special organ of this kind. Beneath
his body, towards the tail, there is a white spot. The
glass shows that this is composed of about a dozen short,
fleshy little tubes, or stumpy fingers, which are some-
times gathered into a cluster, sometimes spread into a
rosette. This bunch of little fingers helps the Glow-
worm to stick to a smooth surface, and also to climb.
If he wishes to fix himself to a pane of glass or a stalk
he opens his rosette, and spreads it wide on the support,
to which it clings by its own natural stickiness. And
by opening and shutting alternately it helps him to creep
along and to climb.
The little fingers that form this rosette are not jointed,
but are able to move in all directions. Indeed they are
more like tubes than fingers, for they cannot seize any-
thing, they can only hold on by their stickiness. They
are very useful, however, for they have a third purpose,
be'sides their powers of clinging and climbing. They
are used as a sponge and brush. At a moment of rest,
[63]
FABRE'S BOOK OF INSECTS
after a meal, the Glow-worm passes and repasses this
brush over his head and sides and his whole body, a per-
formance made possible by the flexibility of his spine.
This is done point by point, from one end of the body
to the other, with a scrupulous care that proves the great
interest he takes in the operation. At first one may
wonder why he should dust and polish himself so care-
fully. But no doubt, by the time he has turned the Snail
into gruel inside the shell and has then spent several
days in eating the result of his labours, a wash and brush-
up is not amiss.
Ill
HIS LAMP
If the Glow-worm possessed no other talent than that
of chloroforming his prey by means of a few tweaks as
gentle as kisses, he would be unknown to the world in
general. But he also knows how to light himself like
a lantern. He shines; which is an excellent manner of
becoming famous.
In the case of the female Glow-worm the lighting-
apparatus occupies the last three divisions of the body.
On each of the first two it takes the form, on the under
surface, of a wide belt of light; on the third division
or segment the bright part is much smaller, and consists
[64]
THE GLOW-WORM
only of two spots, which shine through the back, and are
visible both above and below the animal. From these
belts and spots there comes a glorious white light, deli-
cately tinged with blue.
The male Glow-worm carries only the smaller of these
lamps, the two spots on the end segment, which are
possessed by the entire tribe. These luminous spots
appear upon the young grub, and continue throughout
life unchanged. And they are always visible both on
the upper and lower surface, whereas the two large belts
peculiar to the female shine only below the body.
I have examined the shining belt under the micro-
scope. On the skin a sort of whitewash is spread, formed
of some very fine grain-like substance, which is the source
of the light. Close beside it is a curious air-tube, with
a short wide stem leading to a kind of bushy tuft of
delicate branches. These branches spread over the sheet
of shining matter, and sometimes dip into it.
It is plain to me that the brightness is produced by
the breathing-organs of the Glow-worm. There are
certain substances which, when mixed with air, become
luminous or even burst into flame. Such substances are
called combustible^ and the act of their producing light
or flame by mingling with the air is called oxidisation.
The lamp of the Glow-worm is the result of oxidisation.
The substance that looks like whitewash is the matter
[65]
FABRE'S BOOK OF INSECTS
that is oxidised, and the air is supplied by the tube con-
nected with the Glow-worm's breathing-organs. But
as to the nature of the shining substance, no one as yet
knows anything.
We are better informed as regards another question.
We know that the Glow-worm has complete control of
the light he carries. He can turn it up or down, or
out, as he pleases.
If the flow of air through the tube be increased, the
light becomes more intense: if the same air-tube, in-
fluenced by the will of the animal, stops the passage of'
air, the light grows fainter or even goes out.
Excitement produces an effect upon the air-tube. I
am speaking now of the modest fairy-lamp, the spots
on the last segment of the Glow-worm's body. These
are suddenly and almost completely put out by any kind
of flurry. When I am hunting for young Glow-worms
I can plainly see them glimmering on the blades of grass;
but should the least false step disturb a neighbouring
twig, the light goes out at once and the insect becomes
invisible.
The gorgeous belts of the females, however, are very
little, if at all, affected by even the most violent sur-
prise. I fire a gun, for instance, beside a wire-gauze
cage in which I am rearing a menagerie of female Glow-
worms in the open air. The explosion produces no
[66]
THE GLOW-WORM
result : the illumination continues, as bright and placid
as before. I take a spray, and rain down a slight shower
of cold water upon the flock. Not one of my animals
puts out its light; at the very most there is a brief pause
in the radiance, and then only in some cases. I send
a puff of smoke from my pipe into the cage. This time
the pause is more marked. There are even some lamps
put out, but they are soon relit. Calm returns, and the
light is as bright as ever. I take some of the captives
in my fingers and tease them a little. Yet the illumina-
tion is not much dimmed, if I do not press too hard with
my thumb. Nothing short of very serious reasons would
make the insect put out its signals altogether.
All things considered, there is not a doubt but that
the Glow-worm himself manages his lighting-apparatus,
extinguishing and rekindling it at will; but there is one
circumstance over which the insect has no control. If
I cut off a strip of the skin, showing one of the luminous
belts, and place it in a glass tube, it will shine away
merrily, though not quite as brilliantly as on the living
body. The presence of life is unnecessary, because the
luminous skin is in direct contact with the air, and the
flow of oxygen through the air-tube is therefore not re-
quired. In aerated water the skin shines as brightly as
in the free air, but the light is extinguished in water that
has been deprived of its air by boiling. There could be
[67]
FABRE'S BOOK OF INSECTS
no better proof that the Glow-worm's light is the effect
of oxidisation.
The light is white, calm, and soft to the eyes, and
suggests a spark dropped by the full moon. In spite of
its splendour it is very feeble. If we move a Glow-
worm along a line of print, in perfect darkness, we can
easily make out the letters one by one, and even words
when they are not too long; but nothing is visible beyond
this very narrow zone. A lantern of this kind soon
tires the reader's patience.
These brilliant creatures know nothing at all of family
affection. They lay their eggs anywhere, or rather strew
them at random, either on the earth or on a blade of grass.
Then they pay no further attention to them.
From start to finish the Glow-worm shines. Even
the eggs are luminous, and so are the grubs. At the
approach of cold weather the latter go down into the
ground, but not very far. If I dig them up I find them
with their little stern-lights still shining. Even below
the soil they keep their lanterns bravely alight.
[68]
CHAPTER VI
A MASON-WASP
I
HER CHOICE OF A BUILDING-SITE
OF the various insects that like to make their
home in our houses, certainly the most inter-
esting, for her beautiful shape, her curious
manners, and her wonderful nest, is a certain Wasp
called the Pelopaeus. She is very little known, even to
the people by whose fireside she lives. This is owing
to her quiet, peaceful ways; she is so very retiring that
her host is nearly always ignorant of her presence. It
is easy for noisy, tiresome, unpleasant persons to make
themselves famous. I will try to rescue this modest
creature from her obscurity.
The Pelopaeus is an extremely chilly mortal. She
pitches her tent under the kindly sun that ripens the
olive and prompts the Cicada's song; and even then she
needs for her family the additional warmth to be found
in our dwellings. Her usual refuge is the peasant's
lonely cottage, with its old fig-tree shading the well in
front of the door. She chooses one exposed to all the
[69]
FABRE'S BOOK OF INSECTS
heat of summers, and if possible possessing a big fire-
place in which a fire of sticks always burns. The cheer-
ful blaze on winter evenings has a great influence upon
her choice, for she knows by the blackness of the chimney
that the spot is a likely one. A chimney that is not
well glazed by smoke gives her no confidence: people
must shiver wth cold in that house.
During the dog-days in July and August the visitor
suddenly appears, seeking a place for her nest. She is
not in the least disturbed by the bustle and movement
of the household: they take no notice of her nor she
of them. She examines — now with her sharp eyes, now
with her sensitive antennae — the corners of the blackened
ceiling, the rafters, the chimney-piece, the sides of the
fireplace especially, and even the inside of the flue.
Having finished her inspection and duly approved of
the site she flies away, soon to return with the pellet of
mud which will form the first layer of the building.
The spot she chooses varies greatly, and often it is a
very curious one. The temperature of a furnace appears
to suit the young Pelopaeus : at least the favourite site
is the chimney, on either side of the flue, up to a height
of twenty inches or so. This snug shelter has its draw-
backs. The smoke gets to the nests, and gives them a
glaze of brown or black like that which covers the stone-
work. They might easily be taken for inequalities in the
[70]
A MASON-WASP
mortar. This is not a serious matter, provided that the
flames do not lick against the nests. That would stew
the young Wasps to death in their clay pots. But the
mother Wasps seems to understand this: she only
places her family in chimneys that are too wide for any-
thing but smoke to reach their sides.
But in spite of all her caution one danger remains.
It sometimes happens, while the Wasp is building, that
the approach to the half-built dwelling is barred to her
for a time, or even for the whole day, by a curtain of
steam or smoke. Washing-days are most risky. From
morning till night the housewife keeps the huge cauldron
boiling. The smoke from the hearth, the steam from
the cauldron and the wash-tub, form a dense mist in front
of the fireplace.
It is told of the Water-Ouzel that, to get back to his
nest, he will fly through the cataract under a mill-weir.
This Wasp is even more daring: with her pellet of mud
in her teeth she crosses the cloud of smoke and disappears
behind it, where she becomes invisible, so thick is the
screen. An irregular chirring sound, the song she sings
at her work, alone betrays her presence. The building
goes on mysteriously behind the cloud. The song ceases,
and the Wasp flies back through the steam, quite un-
harmed. She will face this danger repeatedly all day,
until the cell is built, stored with food, and closed.
[71]
FABRE'S BOOK OF INSECTS
Once and once only I was able to observe a Pelopaeus
at my own fireside; and, as it happened, it was a washing-
day. I had not long been appointed to the Avignon
grammar-school. It was close upon two o'clock, and in
a few minutes the roll of the drum would summon me
to give a scientific lecture to an audience of wool-gather-
ers. Suddenly I saw a strange, agile insect dart through
the steam that rose from the wash-tub. The front part of
its body was very thin, and the back part was very plump,
and the two parts were joined together by a long thread.
It was the Pelopaeus, the first I had seen with observant
eyes.
Being very anxious to become better acquainted with
my visitor, I fervently entreated the household not to
disturb her in my absence. Things went better than I
dared hope. On my return she was still carrying on her
mason's work behind the steam. Being eager to see the
building of the cells, the nature of the provisions, and
the evolution of the young Wasps, I raked the fire so as
to decrease the volume of smoke, and for a good two hours
I watched the mother Wasp diving through the cloud.
Never again, in the forty years that followed, was my
fireplace honoured with such a visit. All the further
information I have gathered was gleaned on the hearths
of my neighbours.
The Pelopseus, it appears, is of a solitary and vagrant
[72]
A MASON-WASP
disposition. She nearly always builds a lonely nest, and
unlike many Wasps and Bees, she seldom founds her
family at the spot where she was reared herself. She is
often found in our southern towns, but on the whole she
prefers the peasant's smoky house to the townsman's
white villa. Nowhere have I seen her so plentiful as in
my village, with its tumble-down cottages burnt yellow
by the sun.
It is obvious that this Wasp, when she so often chooses
the chimney as her abode, is not seeking her own comfort:
the site means work, and dangerous work. She seeks the
welfare of her family. This family, then, must require
a high temperature, such as other Wasps and Bees do not
need.
I have seen a Pelopaeus nest in the engine-room of
a silk- factory, iixed to the ceiling just above the huge
boiler. At this spot the thermometer marked 120
degrees all through the year, except at night and on holi-
days.
In a country distillery I have found many nests, fixed
on anything that came to hand, even a pile of account-
books. The temperature of one of these, quite close to
the still, was 113 degrees. It is plain that this
Wasp cheerfully endures a degree of heat that makes the
oily palm-tree sprout.
A boiler or a furnace she regards as the ideal home, but
[73]
FABRE'S BOOK OF INSECTS
she is quite willing to content herself in any snug corner:
a conservatory, a kitchen-ceiling, the recess of a closed
window, the wall of a cottage bedroom. As to the
foundation on which she fixes her nest, she is entirely
indifferent. As a rule she builds her groups of cells
on stonework or timber; but at various times I have seen
nests inside a gourd, in a fur cap, in the hollow of a brick,
on the side of a bag of oats, and in a piece of lead tubing.
Once I saw something more remarkable still, in a farm
near Avignon. In a large room with a very wide fire-
place the soup for the farm-hands and the food for
the cattle simmered in a row of pots. The labourers used
to come in from the fields to this room, and devour their
meal with the silent haste that comes from a keen
appetite. To enjoy this half -hour comfortably they
would take off their hats and smocks, and hang them on
pegs. Short though this meal was, it was long enough to
allow the Wasps to take possession of their garments.
The inside of a straw hat was recognised as a most useful
building-site, the folds of a smock were looked upon as a
capital shelter; and the work of building started at once.
On rising from the table one of the men would shake his
smock, and another his hat, to rid it of the Wasp's nest,
which was already the size of an acorn.
The cook in that farmhouse regarded the Wasps with
no friendly eye. They dirtied everything, she said.
[74]
A MASON-WASP
Dabs of mud on the ceiling, on the walls, or on the
chimney-piece you could put up with; but it was a very
different matter when you found them on the linen and
the curtains. She had to beat the curtains every day
with a bamboo. And it was trouble thrown away. The
next morning the Wasps began building as busily as ever.
II
HER BUILDING
I sympathised with the sorrows of that farm-cook, but
greatly regretted that I could not take her place. How
gladly I would have left the Wasps undisturbed, even if
they had covered all the furniture with mud I How I
longed to know what the fate of a nest would be, if
perched on the uncertain support of a coat or a curtain I
The nest of the Mason-bee is made of hard mortar, which
surrounds the twig on which it is built, and becomes
firmly fixed to it; but the nest of the Pelopaeus Wasp is a
mere blob of mud, without cement or foundations.
The materials of which it is made are nothing but wet
earth or dirt, picked up wherever the soil is damp
enough. The thin clay of a river-bank is very suitable,
but in my stony country streams are rare. I can, how-
ever, watch the builders at my leisure in my own garden,
when a thin trickle of water runs all day, as it does some-
[75]
FABRE'S BOOK OF INSECTS
times, through the little trenches that are cut in my
vegetable plots.
The Pelopaeus Wasps of the neighbourhood soon be-
come aware of this glad event, and come hurrying up to
take advantage of the precious layer of mud, a rare dis-
covery in the dry season. They scrape and skim the
gleaming, shiny surface with their mandibles while
standing high on their legs, with their wings quivering
and their black bodies upraised. No neat little house-
wife, with skirts carefully tucked up out of the dirt,
could be more skilful in tackling a job likely to soil her
clothes. These mud-gatherers have not an atom of dirt
upon them, so careful are they to tuck up their skirts in
their own fashion, that is to say, to keep their whole
body out of the way, all but the tips of their legs and the
busy points of the mandibles with which they work.
In this way a dab of mud is collected, almost the size
of a pea. Taking the load in its teeth the insect flies off,
adds a layer to its building, and soon returns to collect
another pellet. The same method is pursued as long as
the earth remains sufficiently wet, during the hottest
hours of the day.
But the favourite spot is the great fountain in the
village, where the people come to water their mules.
Here there is a constant sheet of black mud which neither
the hottest sunshine nor the strongest wind can dry.
[76]
A MASON-WASP
This bed of mire is very unpleasant for the passers-by,
but the Pelopaeus loves to gather her pellets here, amid
the hoofs of the mules.
Unlike some builders in clay, such as the Mason-bees,
the Wasp does not improve the mud to make it into
mortar, but uses it just as it is. Consequently her nests
are flimsy work, absolutely unfitted to stand the changes
and chances of the open air. A drop of water laid upon
their surface softens the spot touched and reduces it to
mud again, while a sprinkling equal to an average shower
turns it to pap. They are nothing but dried slime, and
become slime again as soon as they are wetted.
It is plain, then, that even if the young Pelopaeus were
not so chilly by nature, a shelter is indispensable for the
nests, which would go to pieces at the first shower of
rain. That is why this Wasp is so fond of human dwell-
ings, and especially of the chimney.
Before receiving its final coating, which covers up the
details of the building, the nest has a certain beauty of
its own. It consists of a cluster of cells, sometimes
arranged side by side in a row — which makes it look
rather like a mouth-organ — but more often grouped m
layers placed one above the other. I have sometimes
counted as many as fifteen cells; some nests contain only
ten; others are reduced to three or four, or even only one.
In shape the cells are not far from cylinders, slightly
[77]
FABRE'S BOOK OF INSECTS
larger at the mouth than at the base. They are a little
more than an inch long, and about half an inch wide.
Their delicate surface is carefully polished, and shows
a series of string-like projections, running cross-wise,
not unlike the twisted cords of some kinds of gold-lace.
Each of these strings is a layer of the building; it comes
from the clod of mud used for the coping of the part
already built. By counting them you can tell how
many journeys the Wasp has made in the course of her
work. There are usually between fifteen and twenty.
For one cell, therefore, the industrious builder fetches
materials something like twenty times.
The mouth of the cells is, of course, always turned
upwards. A pot cannot hold its contents if it be upside
down. And the Wasp's cell is nothing but a pot in-
tended to hold the store of food, a pile of small Spiders.
The cells — built one by one, stuffed full of Spiders,
and closed as the eggs are laid — preserve their pretty
appearance until the cluster is considered large enough.
Then, to strengthen her work, the Wasp covers the whole
with a casing, as a protection and defence. She lays
on the plaster without stint and without art, giving it
none of the delicate finishing-touches which she lavishes
on the cells. The mud is applied just as it is brought,
and merely spread with a few careless strokes. The
beauties of the building all disappear under this ugly
[78]
A MASON-WASP
husk. In this final state the nest is like a great splash
of mud, flung against the wall by accident.
Ill
HER PROVISIONS
Now that we know what the provision-jar is like, we
must find out what it contains.
The young Pelopaeus is fed on Spiders. The food
does not lack variety, even in the same nest and the
same cell, for any Spider may form a meal, as long as
it is not too large for the jar. The Cross Spider, with
three crosses of white dots on her back, is the dish that
occurs oftenest. I think the reason for this is simply
that the Wasp does not go far from home in her hunting-
trips, and the Spider with the crosses is the easiest to
find.
The Spider, armed with poison-fangs, is a dangerous
prey to tackle. When of fair size, she could only be
conquered by a greater amount of daring and skill than
the Wasp possesses. Moreover, the cells are too small
to hold a bulky object. The Wasp, therefore, hunts
game of moderate size. If she meets with a kind of
Spider that is apt to become plump, she always chooses
a young one. But, though all are small, the size of her
victims varies enormously, and this variation in size
[79]
FABRE'S BOOK OF INSECTS
leads also to variation in number. One cell will con-
tain a dozen Spiders, while in another there are only five
or six.
Another reason for her choice of small Spiders is that
she kills them before potting them in her cells. She
falls suddenly upon her prey, and carries it off almost
without pausing in her flight. The skilful paralysis
practised by some insects is unknown to her. This means
that when the food is stored it soon decays. Fortunately
the Spiders are small enough to be finished at a single
meal. If they were large and could only be nibbled
here and there, they would decay, and poison the grubs
in the nest.
I always find the egg, not on the surface of the heap,
but on the first Spider that was stored. There is no
exception to this rule. The Wasp places a Spider at
the bottom of the cell, lays her egg upon it, and then
piles the other Spiders on the top. By this clever plan
the grub is obliged to begin on the oldest of the dead
Spiders, and then go on to the more recent. It always
finds in front of it food that has not had time to decom-
pose.
The egg is always laid on the same part of the Spider,
the end containing the head being placed on the plumpest
spot. This is very pleasant for the grub, for the moment
it is hatched it can begin eating the tenderest and nicest
[80]
Printed in Great Britain
A MASON-WASP
food in the store. Not a mouthful is wasted, however, by
these economical creatures. When the meal is finished
there is practically nothing left of the whole heap of
Spiders. This life of gluttony lasts for eight or ten
days.
The grub then sets to work to spin its cocoon, a sack
of pure, perfectly white silk, extremely delicate. Some-
thing more is required to make this sack tough enough
to be a protection, so the grub produces from its body
a sort of liquid varnish. As soon as it trickles into the
meshes of the silk this varnish hardens, and becomes a
lacquer of exquisite daintiness. The grub then fixes
a hard plug at the base of the cocoon to make all secure.
When finished, the work is amber-yellow, and rather
reminds one of the outer skin of an onion. It has the
same fine texture, the same colour and transparency;
and like the onion skin it rustles when it is fingered.
From it, sooner or later according to temperature, the
perfect insect is hatched.
It is possible, while the Wasp is storing her cell, to
play her a trick which will show how purely mechanical
her instincts are. A cell has just been completed, let
us suppose, and the huntress arrives with her first Spider.
She stores it away, and at once fastens her egg on the
plumpest part of its body. She sets out on a second
[81]
FABRE'S BOOK OF INSECTS
trip. I take advantage of her absence to remove with
my tweezers from the bottom of the cell both the dead
Spider and the egg.
The disappearance of the egg must be discovered by
the Wasp, one would think, if she possesses the least
gleam of intelligence. The egg is small, it is true, but
it lies on a comparatively large object, the Spider. What
will the Wasp do when she finds the cell empty'? Will
she act sensibly, and repair her loss by laying a second
egg? Not at all ; she behaves most absurdly.
What she does is to bring a second Spider, which she
stores away with as much cheerful zeal as if nothing
unfortunate had occurred. She brings a third and a
fourth, and still others, each of whom I remove during
her absence; so that every time she returns from the
chase the storeroom is found empty. I have seen her
persist obstinately for two days in seeking to fill the
insatiable jar, while my patience in emptying it was
equally unflagging. With the twentieth victim — pos-
sibly owing to the fatigue of so many journeys — the
huntress considered that the pot was sufficiently supplied,
and began most carefully to close the cell that contained
absolutely nothing.
The intelligence of insects is limited everywhere in
this way. The accidental difficulty which one insect is
powerless to overcome, any other, no matter what its
[82]
A MASON- WASP
species, will be equally unable to cope with. I could
give a host of similar examples to show that insects are
absolutely without reasoning power, notwithstanding the
wonderful perfection of their work. A long series of
experiments has forced me to conclude that they arc
neither free nor conscious in their industry. They build,
weave, hunt, stab, and paralyse their prey, in the same
way as they digest their food, or secrete the poison of
their sting, without the least understanding of the means
or the end. They are, I am convinced, completely
ignorant of their own wonderful talents.
Their instinct cannot be changed. Experience does
not teach it; time does not awaken a glimmer in its
unconsciousness. Pure instinct, if it stood alone, would
leave the insect powerless in the face of circumstances.
Yet circumstances are always changing, the unexpected
is always happening. In this confusion some power is
needed by the insect — as by every other creature — to
teach it what to accept and what to refuse. It requires
a guide of some kind, and this guide it certainly pos-
sesses. Intelligence is too fine a word for it : I will call
\t discernment.
Is the insect conscious of what it does? Yes, and no.
No, if its action is guided by instinct. Yes, if its action
is the result of discernment.
The Pelopaeus, for instance, builds her cells with earth
[83]
FABRE'S BOOK OF INSECTS
already softened into mud. This is instinct. She has
always built in this way. Neither the passing ages nor
the struggle for life will induce her to imitate the Mason-
bee and make her nest of dry dust and cement.
This mud nest of hers needs a shelter against the rain.
A hiding-place under a stone, perhaps, sufficed at first.
But when she found something better she took possession
of it. She installed herself in the home of man. This
is discernment.
She supplies her young with food in the form of
Spiders. This is instinct, and nothing will ever per-
suade her that young Crickets are just as good. But
should there be a lack of her favourite Cross Spider she
will not leave her grubs unfed; she will bring them other
Spiders. This is discernment.
In this quality of discerment lies the possibility of
future improvement for the insect.
IV
HER ORIGIN
The Pelopaeus sets us another problem. She seeks
the warmth of our fireplaces. Her nest, built of soft
mud which would be reduced to pulp by damp, must
have a dry shelter. Heat is a necessity to her.
Is it possible that she is a foreigner'? Did she come,
[84]
A MASON-WASP
perhaps, from the shores of Africa, from the land of
dates to the land of olives'? It would be natural, in that
case, that she should find our sunshine not warm enough
for her, and should seek the artificial warmth of the fire-
side. This would explain her habits, so unlike those
of the other Wasps, by all of whom mankind is avoided.
What was her life before she became our guest ^
Where did she lodge before there were any houses'?
Where did she shelter her grubs before chimneys were
thought of?
Perhaps, when the early inhabitants of the hills near
Serignan were making weapons out of flints, scraping
goatskins for clothes, and building huts of mud and
branches, those huts were already frequented by the
Pelopseus. Perhaps she built her nest in some bulging
pot, shaped out of clay by the thumbs of our ancestors;
or in the folds of the garments, the skins of the Wolf
and the Bear. When she made her home on the rough
walls of branches and clay, did she choose the nearest
spot, I wonder, to the hole in the roof by which the smoke
was let out'? Though not equal to our chimnevs it may
have served at a pinch.
If the Pelopaeus really lived here with the earliest
human inhabitants, what improvements she has seen I
She too must have profited greatly by civilisation: she
has turned man's increasing comfort into her own.
[85]
FABRE'S BOOK OF INSECTS
When the dwelling with a roof and a ceiling was
planned, and the chimney with a flue was invented, we
can imagine the chilly creature saying to herself :
"How pleasant this is! Let us pitch our tent here."
But we will go back further still. Before huts
existed, before the niche in the rut, before man himself
had appeared, where did the Pelopaeus build'? The
question does not -stand alone. Where did the Swallow
and the Sparrow build before there were windows and
chimneys to build in?
Since the Swallow, the Sparrow, and the Wasp existed
before man, their industry cannot be dependent on the
works of man. Each of them must have had an art of
building in the time when man was not here.
For thirty years and more I asked myself where the
Pelopaeus lived in those times. Outside our houses I
could find no trace of her nests. At last chance, which
favours the persevering, came to my help.
The Serignan quarries are full of broken stones, of
refuse that has been piled there in the course of cen-
turies. Here the Fieldmouse crunches his olive-stones
and acorns, or now and then a Snail. The empty Snail-
shells lie here and there beneath a stone, and within
them different Bees and Wasps build their cells. In
searching for these treasures I found, three times, the
nest of a Pelopaeus among the broken stones.
[86]
A MASON-WASP
These three nests were exactly the same as those
found in our houses. The material was mud, as always;
the protective covering was the same mud. The dangers
of the site had suggested no improvements to the builder.
We see, then, that sometimes, but very rarely, the Pe-
lopaeus builds in stoneheaps and under flat blocks of stone
that do not touch the ground. It was in such places
as these that she must have made her nest before she
invaded our houses.
The three nests, however, were in a piteous state.
The damp and exposure had ruined them, and the cocoons
were in pieces. Unprotected by their earthen cover the
grubs had perished — eaten by a Fieldmouse or another.
The sight of these ruins made me wonder if my neigh-
bourhood were really a suitable place for the Pelopaeus
to build her nest out of doors. It is plain that the mother
Wasp dislikes doing so, and is hardly ever driven to such
a desperate measure. And if the climate makes it im-
possible for her to practise the industry of her forefathers
successfully, I think we may conclude that she is a
foreigner. Surely she comes from a hotter and drier
climate, where there is little rain and no snow.
I believe the Pelopaeus is of African origin. Far
back in the past she came to us through Spain and Italy,
and she hardly ever goes further north than the olive-
trees. She is an African who has become a naturalised
[87]
FABRE'S BOOK OF INSECTS
Provengal. In Africa she is said often to nest under
stones, but in the Malay Archipelago we hear of her kins-
woman in houses. From one end of the world to the
other she has the same tastes — Spiders, mud cells, and the
shelter of a man's roof. If I were in the Malay Archi-
pelago I should turn over the stone-heaps, and should
most likely discover a nest in the original position, under
a flat stone.
[88]
CHAPTER VII
THE PSYCHES
I
A WELL-DRESSED CATERPILLAR
IN the springtime, those who have eyes to see may
find a surprise on old walls and dusty roads.
Certain tiny faggots, for no apparent reason, set
themselves in motion and make their way along by
sudden jerks. The lifeless comes to life : the immovable
moves. This is indeed amazing. If we look closer,
however, we shall solve the riddle.
Enclosed within the moving bundle is a fair-sized
Caterpillar, prettily striped with black and white. He
is seeking for food, and perhaps for some spot where
he can turn into a Moth. He hurries along timidly,
dressed in a queer garment of twigs, which completely
covers the whole of him except his head and the front
part of his body, with its six short legs. At the least
alarm he disappears entirely into his case, and does not
budge again. This is the secret of the walking bundle
of sticks. It is a Faggot Caterpillar, belonging to the
group known as the Psyches.
[89]
FABRE'S BOOK OF INSECTS
To protect himself from the weather the chilly, bare-
skinned Psyche builds himself a portable shelter, a
travelling cottage which the owner never leaves until
he becomes a Moth. It is, indeed, something better than
a hut on wheels, with a thatched roof to it: it is more
like a hermit's frock, made of an unusual kind of ma-
terial. In the valley of the Danube the peasant wears
a goatskin cloak fastened with a belt of rushes. The
Psyche wears even rougher raiment than this : he makes
himself a suit of clothes out of sticks. And since this
would be a regular hair-shirt to a skin so delicate as his,
he puts in a thick lining of silk.
In April, on the walls of my chief workshop — my
stony liarmas with its wealth of insect life — I find the
Psyche who will supply me with my most detailed infor-
mation. He is in the torpid state which shows he will
soon become a Moth. It is a good opportunity for
examining his bundle of sticks, or case.
It is a fairly regular object, shaped like a spindle, and
about an inch and a half long. The pieces that compose
it are fixed in front and free at the back. They are
arranged anyhow, and would form rather a poor shelter
against the sun and rain if the hermit had no other
protection than this.
At the first glance it appears like thatch; but thatch
is not an exact description of it, for grain-stems are rarely
[90]
^
f^
THE PSYCHES
found in it. The chief materials are remnants of very
small stalks, light, soft, and rich in pith; next in order
come bits of grass-leaves, scaly twigs from the cypress-
tree, and all sorts of little sticks; and lastly, if the
favourite pieces run short, fragments of dry leaves.
In short the Caterpillar, while preferring pithy pieces,
will use anything he comes across, provided it be light,
very dry, softened by long exposure, and of the right
size. All his materials are used just as they are, with-
out any alterations or sawings to make them the proper
length. He does not cut the laths that form his roof;
he gathers them as he finds them. His work is limited
to fixing them at the fore-end.
In order to lend itself to the movements of the travel-
ling Caterpillar, and particularly to enable the head and
legs to move freely while a new piece is being fixed in
position, the front part of this case or sheath must be
made in a special way. Here a casing of sticks is no
longer suitable, for their length and stiffness would ham-
per the workman and even make his work impossible.
What is required here is a flexible neck, able to move in
all directions. The collection of stakes, therefore, ends
suddenly at some distance from the fore-part, and is
there replaced by a collar where the silk lining is merely
hardened with very tiny particles of wood, which
strengthen the material without making it less flexible.
[9>]
FABRE'S BOOK OF INSECTS
This collar, which allows of free movement, is so impor-
tant that all the Psyches use it, however greatly the rest
of their work may differ. All carry, in front of the bundle
of sticks, a yielding neck, soft to the touch, formed inside
of a web of pure silk and coated outside with a velvety
sawdust, which the Caterpillar obtains by crushing up
any sort of dry straw.
The same kind of velvet, but dull and faded —
apparently through age — finishes the sheath at the back,
in the form of a rather long projection, open at the end.
When I remove the outside of the straw casing, shred-
ding it piece by piece, I find a varying number of laths,
or tiny sticks. I have counted as many as eighty, and
more. Underneath it I find, from one end of the Cater-
pillar to the other, the same kind of inner sheath that
was formerly visible at the front and back only. This
inner sheath is composed everywhere of very strong silk,
which resists without breaking when pulled by the
fingers. It is a smooth tissue, beautifully white inside,
drab and wrinkled outside, where it bristles with a crust
of woody particles.
Later on we shall see how the Caterpillar makes him-
self this complicated garment, formed of three layers,
one placed upon the other in a definite order. First
comes the extremely fine satin which is in direct contact
[92]
THE PSYCHES
with the skin; next, the mixed stuff dusted with woody
matter, which saves the silk and gives strength to the
work; and lastly the outer casing of overlapping sticks.
Although all the Psyches wear this threefold garment,
the different species make distinct variations in the outer
case. There is one kind, for instance, whom I am apt
to meet towards the end of June, hurrying across some
dusty path near the houses. His case surpasses that of
the first species, both in size and in regularity of arrange-
ment. It forms a thick coverlet of many pieces, in which
I recognise fragments of hollow stalks, bits of fine straw,
and perhaps 'blades of grass. In front there is never
any flounce of dead leaves, a troublesome piece of finery
which is pretty frequent, though not always used, in the
costume of the first species I described. At the back
there is no long projection beyond the outer covering.
Save for the indispensable collar at the neck, the whole
Caterpillar is cased in sticks. There is not much variety
about the thing, but, when all is said, there is a certain
beauty in its stern faultlessness.
There is a smaller and more simply dressed Psyche
who is very common at the end of winter on the walls,
as well as in the bark of gnarled old trees, whether olive-
trees or elms, or indeed almost any other. His case, a
modest little bundle, is hardly more than two-fifths of
[93]
FABRE'S BOOK OF INSECTS
an inch in length. A dozen rotten straws, picked up
at random and fixed close to one another in a parallel
direction, represent, with the silk sheath, his whole out-
lay on dress.
It would be difficult to clothe oneself more economi-
cally.
II
A DEVOTED MOTHER
If I gather a number of little Psyches in April and
place them in a wire bell-jar, I can find out more about
them. Most of them are in the chrysalis state, waiting
to be turned into Moths, but a few are still active and
clamber to the top of the wire trellis. There they fix
themselves by means of a little silk cushion, and both
they and I must wait for weeks before anything further
happens.
At the end of June the male Psyche comes out of his
case, no longer a Caterpillar, but a Moth. The case,
or bundle of sticks, you will remember, had two openings,
one in front and one at the back. The front one, which
is the more regular and carefully made, is permanently
closed by being fastened to the support on which the
chrysalis is fixed; so the Moth, when he is hatched, is
obliged to come out by the opening at the back. The
[94]
THE PSYCHES
Caterpillar turns round inside the case before he changes
into a Moth.
Though they wear but a simple pearl-grey dress and
have insignificant wings, hardly larger than those of a
Common Fly, these little male Moths are graceful
enough. They have handsome feathery plumes for
antennae, and their wings are edged with delicate fringes.
For the appearance of the female Psyche, however,
little can be said.
Some days later than the others she comes out of the
sheath, and shows herself in all her wretchedness. Call
that little fright a Moth! One cannot easily get used
to the idea of so miserable a sight : as a Caterpillar she
was no worse to look at. There are no wings, none at
all ; there is no silky fur either. At the tip of her round,
tufty body she wears a crown of dirty-white velvet; on
each segment, in the middle of the back, is a large, rec-
tangular, dark patch — her sole attempts at ornament.
The mother Psyche renounces all the beauty which her
name of Moth seems to promise.
As she leaves her chrysalid sheath she lays her eggs
within it, thus bequeathing the maternal cottage (or the
maternal garment, if you will) to her heirs. As she lays
a great many eggs the affair takes some thirty hours.
When the laying is finished she closes the door and makes
everything safe against invasion. For this purpose
[95]
FABRE'S BOOK OF INSECTS
some kind of wadding is required. The fond mother
makes use of the only ornament which, in her extreme
poverty, she possesses. She wedges the door with the
coronet of velvet which she carries at the tip of her body.
Finally she does even more than this. She makes a
rampart of her body itself. With a convulsive move-
ment she dies on the threshold of her recent home, her
cast chrysalid skin, and there her remains dry up. Even
after death she stays at her post.
If the outer case be now opened it will be found to
contain the chrysalid wrapper, uninjured except for the
opening in front, by which the Psyche came out. The
male Moth, when obliged to make his way through the
narrow pass, would find his wings and his plumes very
cumbersome articles. For this reason he makes a start
for the door while he is still in the chrysalis state, and
comes half-way out. Then, as he bursts his amber-
coloured tunic, he finds, right in front of him, an open
space where flight is possible.
But the mother Moth, being unprovided with wings
and plumes, is not compelled to take any such precau-
tions. Her cylinder-like form is bare, and differs very
little from that of the Caterpillar. It allows her to
crawl, to slip into the narrow passage, and to come forth
without difficulty. So she leaves her cast skin behind
[96]
THE PSYCHES
her, right at the back of the case, well covered by the
thatched roof.
And this is an act of prudence, showing her deep
concern for the fate of her eggs. They are, in fact,
packed as though in a barrel, in the parchment-like bag
formed by the cast skin. The Moth has methodically
gone on laying eggs in that receptacle till it is full. Not
satisfied with bequeathing her house and her velvet
coronet to her offspring, as the last act of her life she
leaves them her skin.
Wishing to observe the course of events at my ease
I once took one of these chrysalid bags, stuffed with eggs,
from its outer casing of sticks, and placed it by itself,
beside its case, in a glass tube. In the first week of
July I suddenly found myself in possession of a large
family. The hatching took place so quickly that the
new-born Caterpillars, about forty in number, had
already clothed themselves in my absence.
They wore a garment like a sort of Persian head-dress,
in dazzling white plush. Or, to be more commonplace,
a white cotton night-cap without a tassel. Strange to
say, however, instead of wearing their caps on their
heads, they wore them standing up from their hind-
quarters, almost perpendicularly. They roamed about
gaily inside the tube, which was a spacious dwelling for
[97]
FABRE'S BOOK OF INSECTS
such mites. I was quite determined to find out with
what materials and in what manner the first outlines
of the cap were woven.
Fortunately the chrysalid bag was far from being
empty. I found within the rumpled wrapper a second
family as numerous as those already out of the case.
Altogether there must have been five or six dozen eggs.
I transferred to another place the little Caterpillars who
were already dressed, keeping only the naked new-comers
in the tube. They had bright red heads ; the rest of their
bodies was dirty-white; and they measured hardly a
twenty-fifth of an inch in length.
I had not long to wait. The next day, little by little,
singly or in groups, the little laggards left the chrysa-
lid bag. They came out without breaking that frail
object, through the opening in front made by their
mother. Not one of them used it as a dress-material,
though it had the delicacy and amber colouring of an
onion-skin ; nor did any of them make use of a certain fine
quilting that lines the inside of the bag and forms an ex-
quisitely soft bed for the eggs. One would have thought
this downy stuff would make an excellent blanket for the
chilly creatures, but not a single one used it. There
would not be enough to go round.
They all went straight to the coarse outer casing of
sticks, which I had left in contact with the chrysalid skin
[98]
THE PSYCHES
containing the eggs. The matter was urgent, they
evidently felt. Before making your entrance into the
world and going a-hunting, you must first be clad. All
therefore, with equal fury, attacked the old sheath and
hastily dressed themselves in their mother's old clothes.
Some turned their attention to bits that happened to be
opened lengthwise, scraping the soft white inner layer;
others, greatly daring, penetrated into the tunnel of a
hollow stalk and collected their materials in the dark.
The courage of these was rewarded; they secured first-
rate materials and wove garments of dazzling white.
There were others who bit deeply into the piece they
chose, and made themselves a motley covering, in which
the snOwy whiteness was marred by darker particles.
The tools the little Caterpillars use for this purpose are
their mandibles, which are shaped like wide shears and
have five strong teeth apiece. The two blades fit into
each other, and form an instrument capable of seizing
and slicing any fibre, however small. Under the micro-
scope it is seen to be a wonderful specimen of mechanical
precision and power. If the Sheep had a similar tool in
proportion to her size, she could browse on the stems of
trees instead of the grass.
It is very instructive to watch these Psyche-grubs toil-
ing to make themselves a cotton night-cap. There are
numbers of things to remark, both in the finish of the work
[99]
C^Sfo'oVf:
FABRE'S BOOK OF INSECTS
and the skill of the methods they employ. They are so
tiny that while I observe them through my magnifying
glass I must be careful not to breathe, lest I should over-
turn them or puff them away. Yet this speck is expert in
the art of blanket-making. An orphan, born but a
moment ago, it knows how to cut itself a garment out of
its mother's old clothes. Of its methods I will tell you
more presently, but first I must say another word with
regard to its dead mother.
I have spoken of the downy quilting that covers the in-
side of the chrysalid bag. It is like a bed of eider-down,
on which the little Caterpillars rest for a while after leav-
ing the egg. Warmly nestling in this soft rug they pre-
pare themselves for their plunge into the outer world of
work.
The Eider robs herself of her down to make a luxurious
bed for her brood; the mother Rabbit shears from her own
body the softest part of her fur to provide a mattress for
her new-born family. And the same thing is done by the
Psyche.
The mass of soft wadding that makes a warm coverlet
for the baby Caterpillar is a material of incomparable
delicacy. Through the microscope it can be recognised as
the scaly dust, the intensely fine down in which every
Moth is clad. To give a snug shelter to the little grubs
who will soon be swarming in the case, to provide them
[100]
THE PSYCHES
with a refuge in which they can play about and gather
strength before entering the wide world, the Psyche
strips herself of her fur like the mother Rabbit.
This may possibly be done mechanically; it may be the
unintentional effect of rubbing repeatedly against the
low-roofed walls ; but there is nothing to tell us so. Even
the humblest mother has her foresight. It is quite likely
that the hairy Moth twists about, and goes to and fro in
the narrow passage, in order to get rid of her fleece and
prepare bedding for her family.
I have read in books that the young Psyches begin life
by eating up their mother. I have seen nothing of the
sort, and I do not even understand how the idea arose.
Indeed, she has given up so much for her family that
there is nothing left of her but some thin, dry strips — not
enough to provide a meal for so numerous a brood. No,
my little Psyches, you do not eat your mother. In vain
do I watch you : never, either to clothe or to feed him-
self, does any one of you lay a tooth upon the remains of
the deceased.
Ill
A CLEVER TAILOR
I will now describe in greater detail the dressing of
the grubs.
[101]
FABRE'S BOOK OF INSECTS
The hatching of the eggs takes place in the first fort-
night of July. The head and upper part of the little
grubs are of a glossy black, the next two segments are
brownish, and the rest of the body is a pale amber. They
are sharp, lively little creatures, who run about with
short, quick steps.
For a time, after they are out of the bag where they are
hatched, they remain in the heap of fluif that was stripped
from their mother. Here there is more room, and more
comfort too, than in the bag whence they came ; and while
some take a rest, others bustle about and exercise them-
selves in walking. They are all picking up strength be-
fore leaving the outer case.
They do not stay long amid this luxury. Gradually,
as they gain vigour, they come out and spread over the
surface of the case. Work begins at once, a very urgent
work — that of dressing themselves. By and by they
will think of food: at present nothing is of any import-
ance but clothes.
Montaigne, when putting on a cloak which his father
had worn before him, used to say, "I dress myself in my
father." Well, the young Psyches in the same way dress
themselves in their mother. (In the same way, it must
be remembered; not in her skin, but in her clothes.)
From the outer case of sticks, which I have sometimes
described as a house and sometimes as a garment, they
[102]
THE PSYCHES
scrape the material to make themselves a frock. The
stuff they use is the pith of the little stalks, especially of
the pieces that are split lengthwise, because the contents
are more easily taken from these.
The manner of beginning the garment is worth noting.
The tiny creature employs a method as ingenious as
any that we could hope to discover. The wadding is
collected in pellets of infinitesimal size. How are these
little pellets to be fixed and joined together *? The
manufacturer needs a support, a base; and this support
cannot be obtained on the Caterpillar's own body. The
difficulty is overcome very cleverly. The pellets are
gathered together, and by degrees fastened to one another
with threads of silk — for the Caterpillar, as you know,
can spin silk from his own body as the Spider spins her
web. In this way a sort of garland is formed, with the
pellets or particles swinging in a row from the same rope.
When it is long enough this garland is passed round
the waist of the little creature, in such a way as to leave
its six legs free. Then it ties the ends together with a
bit of silk, so that it forms a girdle round the grub's
body.
This girdle is the starting-point and support of the
whole work. To lengthen it, and enlarge it into a com-
plete garment, the grub has only to fix to it the scraps
of pith which the mandibles never cease tearing from
[103]
FABRE'S BOOK OF INSECTS
the case. These scraps or pellets are sometimes placed
at the top, sometimes at the bottom or side, but they are
always fixed at the fore-edge. No device could be better
contrived than this garland, first laid out flat and then
buckled like a belt round the body.
Once this start is made the weaving goes on well.
Gradually the girdle grows into a scarf, a waistcoat, a
short jacket, and lastly a sack, and in a few hours it is
complete — a conical hood or cloak of magnificent white-
ness.
Thanks to his mother's care the little grub is spared
the perils of roaming about in a state of nakedness. If
she did not place her family in her old case they might
have great difficulty in clothing themselves, for straws
and stalks rich in pith are not found everywhere. And
yet, unless they died of exposure, it appears that sooner
or later they would find some kind of garment, since
they seem ready to use any material that comes to hand.
I have made many experiments with new-born grubs
in a glass tube.
From the stalks of a sort of dandelion they scraped,
without the least hesitation, a superb white pith, and
made it into a delicious white cloak, much finer than
any they would have obtained from the remains of their
mother's clothes. An even better garment was woven
from some pith taken from the kitchen-broom. This
[104]
THE PSYCHES
time the work glittered with little sparks, like specks
of crystal or grains of sugar. It was my manufacturers'
masterpiece.
The next material I offered them was a piece of
blotting-paper. Here again my grubs did not hesitate :
they lustily scraped the surface and made themselves a
paper coat. Indeed, they were so much pleased with
this that when I gave them their native case they scorned
it, preferring the blotting-paper.
To others I gave nothing at all. Not to be baffled,
however, they hastened to scrape the cork of the tube
and break it into atoms. Out of these they made them-
selves a frock of cork-grains, as faultless as though they
and their ancestors had always made use of this material.
The novelty of the stuff, which perhaps no Caterpillar
had ever used before, made no difference in the cut of the
garment.
Finding them ready to accept any vegetable matter
that was dry and light, I next tried them with animal
and mineral substances. I cut a strip from the wing of
a Great Peacock Moth, and placed two little naked
Caterpillars upon it. For a long time they both hesi-
tated. Then one of them resolved to use the strange
carpet. Before the day was over he had clothed himself
in grey velvet made of the Great Peacock's scales.
I next took some soft, flaky stones, such as will break
[105]
FABRE'S BOOK OF INSECTS
at the merest touch into atoms nearly as fine as the dust
on a Butterfly's wing. On a bed of this powdery stuff,
which glittered like steel filings, I placed four Cater-
pillars in need of clothes. One, and one alone, decided
to dress himself. His metallic garment, from which
the light drew flashes of every colour of the rainbow,
was very rich and sumptuous, but mightily heavy and
cumbrous. Walking became laborious under that load
of metal. Even so must a Byzantine Emperor have
walked at ceremonies of State.
In cases of necessity, then, the young Caterpillar does
not shrink from acts of sheer madness. So urgent is
his need to clothe himself that he will weave mineral
matter rather than go naked. Food means less to him
than clothes. If I make him fast for a couple of days,
and then, having robbed him of his garment, place him
on his favourite fodd, a leaf of very hairy hawkweed, he
will make himself a new coat before satisfying his
hunger.
This devotion to dress is due, not to any special sensi-
tiveness to cold, but to the young Caterpillar's foresight.
Other Caterpillars take shelter among the leaves, in
underground cells, or in the cracked bark of trees, but
the Psyche spends his winter exposed to the weather.
He therefore prepares himself, from his birth, for the
perils of the cold season.
[106]
THE PSYCHES
As soon as he is threatened with the rains of autumn
he begins to work upon his outer case. It is very rough
at first. Straws of uneven length and bits of dry leaves
are fastened, with no attempt at order, behind the neck
of the sack or undergarment, which must remain flexible
so as to allow the Caterpillar to bend freely in every
direction. These untidy first logs of the outer case will
not interfere with the final regularity of the building:
they will be pushed back and driven out as the sack
grows longer in front.
After a time the pieces are longer and more carefully
chosen, and are all laid on lengthwise. The placing
of a straw is done with surprising speed and skill. The
Caterpillar turns it round and round between his legs,
and then, gripping it in his mandibles, removes a few
morsels from one end, and immediately fixes them to
the end of the sack. He probably does this in order that
the silk may obtain a firmer hold, as a plumber gives a
touch of the file to a point that is to be soldered.
Then, by sheer strength of jaw, he lifts and brandishes
his straw in the air before laying it on his back. At
once the spinneret sets to work and fixes it in place.
Without any groping about or correcting, the thmg is
done. By the time the cold weather arrives the warm
case is complete.
But the silky felt of the interior is never thick enough
[107]
FABRE'S BOOK OF INSECTS
to please the Caterpillar. When spring comes he spends
all his spare time in improving his quilt, in making it
ever thicker and softer. Even if I take ofE his outer case
he refuses to rebuild it : he persists in adding new layers
to the lining, even when there is nothing to be lined.
The sack is lamentably flabby; it sags and rumples. He
has no protection nor shelter. No matter. The hour
for carpentry has passed. The hour has come for up-
holstering; and he upholsters obstinately, padding a
house — or lining a garment — that no longer exists. He
will perish m'iserably, cut up by the Ants, as the result
of his too-rigid instinct.
[108]
CHAPTER VIII
THE SELF-DENIAL OF THE SPANISH COPRIS
YOU remember, I hope, the Sacred Beetle, who
spends her time in making balls, both to
serve as food and also to be the foundation
of her pear-shaped nest. I pointed out the advantages
of this shape for the young Beetles, since the globe is
the best form that could be invented to keep their pro-
visions from becoming dry and hard.
After watching this Beetle at work for a long time I
began to wonder if I had not perhaps been mistaken in
admiring her instinct so greatly. Was it really care
for her grubs, I asked myself, that taught her to provide
them with the tenderest and most suitable food*? It is
the trade of the Sacred Beetle to make balls. Is it won-
derful that she should continue her ball-making under-
ground? A creature built with long curved legs, very
useful for rolling balls across the fields, will go on with
her favourite occupation wherever she may be, without
regard to her grubs. Perhaps the shape of the pear is
mere chance.
To settle this question satisfactorily in my own mind
I should need to be shown a Scavenger Beetle who was
[109]
FABRE'S BOOK OF INSECTS
utterly unfamiliar with the ball-making business in
everyday life, and who yet, when laying-time was at
hand, made an abrupt change in her habits and stored
her provisions in the form of a round lump. That would
show me that it was not merely custom, but care for her
grubs, that made her choose the globular shape for her
nest.
Now in my neighbourhood there is a Beetle of this
very kind. She is one of the handsomest and largest,
though not so imposing as the Sacred Beetle. Her name
is the Spanish Copris, and she is remarkable for the sharp
slope of her chest and the size of the horn surmounting
her head.
Being round and squat, the Spanish Copris is certainly
incapable of such gymnastics as are performed by the
Sacred Beetle. Her legs, which are insignificant in
length, and which she folds under her body at the
slightest alarm, are not in the least like the stilts of the
pill-rollers. Their stunted form and their lack of flexi-
bility are enough in themselves to tell us that their owner
would not care to roam about burdened with a rolling
ball.
The Copris, indeed, is not of an active nature. Once
she has found her provisions, at night or in the evening
twilight, she begins to dig a burrow on the spot. It is
a rough cavern, large enough to hold an apple. Here
[110]
SELF-DENIAL OF THE SPANISH COPRIS
is introduced, bit by bit, the stuff that is just overhead,
or at any rate lying on the threshold of the cave. An
enormous supply of food is stored in a shapless mass,
plain evidence of the insect's gluttony. As long as
the hoard lasts the Copris remains underground. When
the larder is empty the insect searches out a fresh supply
of food, and scoops out another burrow.
For the time being the Copris is merely a scavenger,
a gatherer of manure. She is evidently quite ignorant,
at present, of the art of kneading and modelling a round
loaf. Besides, her short clumsy legs seem utterly un-
suited for any such art.
In May or June, however, comes laying-time. The
insect becomes very particular about choosing the softest
materials for her family's food. Having found what
pleases her, she buries it on the spot, carrying it down
by armfuls, bit by bit. There is no travelling, no cart-
ing, no preparation. I observe, too, that the burrow is
larger and better built than the temporary abodes in
which the Copris takes her own meals.
Finding it difficult to observe the insect closely in its
wild state, I resolved to place it in my insect-house, and
there watch it at my ease.
The poor creature was at first a little nervous in
captivity, and when she had made her burrow was very
cautious about entering it. By degrees, however, she
[111]
FABRE'S BOOK OF INSECTS
was reassured, and in a single night she stored a supply
of the food I had provided for her.
Before a week was out I dug up the soil in my insect-
house, and brought to light the burrow I had seen her
storing with provisions. It was a spacious hall, with
an irregular roof and an almost level floor. In a corner
was a round hole leading to a slanting gallery, which ran
up to the surface of the soil. The walls of this dwelling,
which was hollowed out of fresh earth, had been care-
fully compressed, and were strong enough to resist the
earthquake caused by my experiments. It was easy to
see that the insect had put forth all her skill, all her
digging-powers, in the making of this permanent home,
whereas her own dining-room had been a mere cave, with
walls that were none too safe.
I suspect she is helped, in the building of this archi-
tectural masterpiece, by her mate: at least I often see
him with her in the burrows. I also believe that he lends
his partner a hand with the collecting and storing of the
provisions. It is a quicker job when there are two to
work. But once the home is well stocked he retires:
he makes his way back to the surface and settles down
elsewhere. His part in the family mansion is ended.
Now what do I find in this mansion, into which I
have seen so many tiny loads of provisions lowered?
A mass of small pieces, heaped together anyhow? Not
[1121
SELF-DENIAL OF THE SPANISH COPRIS
a bit of it. I always find a simple lump, a huge mass
which fills the dwelling except for a narrow passage.
This lump has no fixed shape. I come across some
that are like a Turkey's egg in form and size; some the
shape of a conmion onion; I find some that are almost
round, and remind me of a Dutch cheese; I see some
that are circular, with a slight swelling on the upper
surface. In every case the surface is smooth and nicely
curved.
There is no mistaking what has happened. The
mother has collected and kneaded into one lump the
numerous fragments brought down one after the other.
Out of all those particles she has made a single lump,
by mashing them, working them together, and treading
on them. Time after time I have seen her on top of
the colossal loaf which is so much larger than the ball
of the Sacred Beetle — a mere pill in comparison. She
strolls about on the convex surface, which sometimes
measures as much as four inches across; she pats the mass,
and makes it firm and level. I only catch a sight of the
curious scene, for the moment she sees me she slips down
the curved slope and hides away.
With the help of a row of glass jars, all enclosed in
opaque sheaths of cardboard, I can find out a good many
interesting things. In the first place I have found that
the big loaf does not owe its curve — which is always
[113]
FABRE'S BOOK OF INSECTS
regular, no matter how much the slope may vary — to
any rolling process. Indeed I already knew that so
large a mess could not have been rolled into a hole that
it nearly fills. Besides, the strength of the insect would
be unequal to moving so great a load.
Every time I go to the jar the evidence is the same.
I always see the mother Beetle twisted on top of the
lump, feeling here and feeling there, giving little taps,
and making the thing smooth. Never do I catch her
looking as if she wanted to turn the block. It is clear as
daylight that rolling has nothing to do with the matter.
At last it is ready. The baker divides his lump of
dough into smaller lumps, each of which will become a
loaf. The Copris does the same thing. By making
a circular cut with the sharp edge of her forehead, and
at the same time using the saw of her fore-legs, she de-
taches from the mass a piece of the size she requires.
In giving this stroke she has no hesitation: there are
no after-touches, adding a bit here and taking ofE a bit
there. Straight away, with one sharp, decisive cut, she
obtains the proper-sized lump.
Next comes the question of shaping it. Clasping it
as best she can in her short arms, so little adapted, one
would think, for work of this kind, the Copris rounds
her lump of food by pressure, and pressure only. Sol-
[114]
SELF-DENIAL OF THE SPANISH COPRIS
emnly she moves about on the still shapeless mass,
climbs up, climbs down, turns to right and left, above
and below, touching and re-touching with unvarying
patience. Finally, after twenty-four hours of this work,
the piece that was all corners has become a perfect sphere,
the size of a plum. There in her cramped studio, with
scarcely room to move, the podgy artist has completed
her work without once shaking it on its base: by dint of
time and patience she has obtained the exact sphere which
her clumsy tools and her confined space seemed to render
impossible.
For a long time she continues to polish up the globe
with affectionate touches of her foot, but at last she is
satisfied. She climbs to the top, and by simple pressure
hollows out a shallow cavity. In this basin she lays
an egg.
Then, with extreme caution and delicacy, she brings
together the sides of the basin so as to cover the egg, and
carefully scrapes the sides towards the top, which begins
to taper a little and lengthen out. In the end the ball
has become ovoid, or egg-shaped.
The insect next helps herself to a second piece of the
cut loaf, which she treats in the same way. The remain-
der serves for a third ovoid, or even a fourth. The
Sacred Beetle, you remember, made a single pear-shaped
[115]
FABRE'S BOOK OF INSECTS
nest in a way that was familiar to her, and then left her
egg underground while she engaged in fresh enterprises.
The Copris behaves very differently.
Her burrow is almost filled by three or four ovoid
nests, standing one against the other, with the pointed
end upwards. After her long fast one would expect her
to go away, like the Sacred Beetle, in search of food.
On the contrary, however, she stays where she is. And
yet she has eaten nothing since she came underground,
for she has taken good care not to touch the food prepared
for her family. She will go hungry rather than let her
grubs suffer.
Her object in staying is to mount guard over the
cradles. The pear of the Sacred Beetle suffers from the
mother's desertion. It soon shows cracks, and becomes
scaly and swollen. After a time it loses its shape. But
the nest of the Copris remains perfect, owing to the
mother's care. She goes from one to the other, feels
them, listens to them, and touches them up at points
where my eye can detect no flaw. Her clumsy horn-shod
foot is more sensitive in the darkness than my sight in
broad daylight : she feels the least threatening of a crack
and attends to it at once, lest the air should enter and
dry up her eggs. She slips in and out of the narrow
spaces between the cradles, inspecting them with the
utmost care. If I disturb her she sometimes rubs the
[116]
SELF-DENIAL OF THE SPANISH COPRIS
tip of her body against the edge of her wing-cases, making
a soft rustling sound, like a murmur of complaint. In
this way, caring industriously for her cradles, and some-
times snatching a brief sleep beside them the mother
waits.
The Copris enjoys in her underground home a rare
privilege for an insect: the pleasure of knowing her
family. She hears her grubs scratching at the shell to
obtain their liberty; she is present at the bursting of the
nest which she has made so carefully. And when the
little captive, stiffening his legs and humping his back,
tries to split the ceiling that presses down on him, it is
quite possible that the mother comes to his assistance
by making an assault on the nest from the outside.
Being fitted by instinct for repairing and building, why
should she not also be fitted for demolishing"? How-
ever, I will make no assertions, for I have been unable
to see.
Now it is possible to say that the mother Copris,
being imprisoned in an enclosure from which she cannot
escape, stays in the midst of her nest because she has no
choice in the matter. Yet, if this were so, would she
trouble about her work of polishing and constant in-
spection? These cares evidently are natural to her:
they form part of her habits. If she were anxious to
regain her liberty, she would surely roam restlessly round
[117]
FABRE'S BOOK OF INSECTS
the enclosure, whereas I always see her very quiet and
absorbed.
To make certain, I have inspected my glass jars at
different times. She could go lower down in the sand
and hide anywhere she pleased, if rest were what she
wanted; she could climb outside and sit down to fresh
food, if refreshment became necessary. Neither the
prospect of rest in a deeper cave nor the thought of the
sun and of food -makes her leave her family. Until the
last of them has burst his shell she sticks to her post.
I always find her beside her cradles.
For four months she is without food of any kind.
She was no better than a glutton at first, when there was
no family to consider, but now she becomes self-denying
to the point of prolonged fasting. The Hen sitting on
her eggs forgets to eat for some weeks; the watchful
Copris mother forgets food for a third part of the year.
The summer is over. The rains so greatly desired
by man and beast have come at last, soaking the ground
to some depth. After the torrid and dusty days of our
Proven(^al summer, when life is in suspense, we have
the coolness that revives it. The heath puts out its
first pink bells; the autumnal squill lifts its little spike
of lilac flowers; the strawberry-tree's coral bells begin
to soften; the Sacred Beetle and the Copris burst their
[118]
SELF-DENIAL OF THE SPANISH COPRIS
shells, and come to the surface in time to enjoy the last
fine weather of the year.
The newly released Copris family, accompanied by
their mother, gradually emerge from underground.
There are three or four of them, five at most. The
sons are easily recognised by the greater length of their
horns; but there is nothing to distinguish the daughters
from the mother. For that matter, the same confusion
exists among themselves. An abrupt change has taken
place. The mother whose devotion was lately so remark-
able is now utterly indifferent to the welfare of her
family. Henceforward each looks after his own home
and his own interests. They no longer have anything
to do with one another.
The present indifference of the mother Beetle must
not make us forget the wonderful care she has lavished
for four months on end. Except among the Bees,
Wasps, and Ants, who spoon-feed their young and bring
them up with every attention to their health, I know of
no other such case of maternal self-denial. Alone and
unaided she provides each of her children with a cake of
food, whose crust she constantly repairs, so that it be-
comes the safest of cradles. So intense is her affection
that she loses all desire and need of food. In the dark-
ness of the burrow she watches over her brood for four
[119]
FABRE'S BOOK OF INSECTS
months, attending to the wants of the egg, the grub, the
undeveloped Beetle, and the full-grown insect. She
does not return to the glad outer life till all her family
are free. Thus we see one of the most brilliant exam-
ples of maternal instinct in a humble scavenger of the
fields. The Spirit breatheth where He will.
[120]
CHAPTER IX
TWO STRANGE GRASSHOPPERS
I
THE EMPUSA
THE sea, where life first appeared, still preserves
in its depths many of those curious shapes
which were the earliest specimens of the ani-
mal kingdom. But the land has almost entirely lost
the strange forms of other days. The few that remain
are mostly insects. One of these is the Praying Mantis,
whose remarkable shape and habits I have already
described to you. Another is the Empusa.
This insect, in its undeveloped or larval state, is
certainly the strangest creature in all Provence : a slim,
swaying thing of so fantastic an appearance that unaccus-
tomed fingers dare not lay hold of it. The children of
my neighbourhood are so much impressed by its startling
shape that they call it "the Devilkin." They imagine
it to be in some way connected with witchcraft. One
comes across it, though never in great numbers, in the
spring up to May; in autumn; and sometimes in
winter if the sun be strong. The tough grasses of the
[121]
FABRE'S BOOK OF INSECTS
waste-lands, the stunted bushes which catch the sun-
shine and are sheltered from the wind by a few heaps of
stones, are the chilly Empusa's favourite dwelling.
I will tell you, as well as I can, what she looks like.
The tail-end of her body is always twisted and curved
up over her back so as to form a crook, and the lower
surface of her body (that is to say, of course, the upper
surface of the crook) is covered with pointed, leaf-shaped
scales, arranged in three rows. The crook is propped
on four long, thin legs, like stilts ; and on each of these
legs, at the point where the thigh joins the shin, is a
curved, projecting blade not unlike that of a cleaver.
In front of this crook on stilts, this four-legged stool,
there rises suddenly — very long and almost perpendicu-
lar— the stiff corselet or bust. It is round and slender
as a straw, and at the end of it is the hunting-trap, copied
from that of the Mantis. This consists of a harpoon
sharper than a needle, and a cruel vice with jaws toothed
like a saw. The jaw, or blade formed by the upper arm,
is hollowed into a groove and carries five long spikes
on each side, with smaller indentations in between.
The jaw formed by the fore-arm is grooved in the same
way, but the teeth are finer, closer, and more regular.
When at rest, the saw of the fore-arm fits into the groove
of the upper arm. If the machine were only larger it
would be a fearful instrument of torture.
[122]
TWO STRANGE GRASSHOPPERS
The head is in keeping with this arsenal. What a
queer head it is I A pointed face, with curled mous-
taches; large .goggle eyes; between them the blade of a
dirk; and on the forehead a mad, unheard-of thing — a
sort of tall mitre, an extravagant head-dress that juts
forward, spreading right and left into peaked wings.
What does the Devilkin want with that monstrous
pointed cap, as magnificent as any ever worn by astrol-
oger of old^ The use of it will appear presently.
The creature's colouring at this time is commonplace —
chiefly grey. As it develops it becomes faintly striped
with pale green, white, and pink.
If you come across this fantastic object in the bramble-
bushes, it sways upon its four stilts, it wags its head
at you knowingly, it twists its mitre round and peers
over its shoulder. You seem to see mischief in its
pointed face. But if you try to take hold of it this
threatening attitude disappears at once; the raised
corselet is lowered, and the creature makes off with
mighty strides, helping itself along with its weapons,
with which it clutches the twigs. If you have a prac-
ticed eye, however, the Empusa is easily caught, and
penned in a cage of wire-gauze.
At first I was uncertain how to feed them. My Devil-
kins were very little, a month or two old at most. I gave
them Locusts suited to their size, the smallest I
[123]
FABRE'S BOOK OF INSECTS
could find. They not only refused them, but were
afraid of them. Any thoughtless Locust that meekly
approached an Empusa met with a bad reception. The
pointed mitre was lowered, and an angry thrust sent the
Locust rolling. The wizard's cap, then, is a defensive
weapon. As the Ram charges with his forehead, so the
Empusa butts with her mitre.
I next offered her a live House-fly, and this time the
dinner was accepted at once. The moment the Fly came
within reach the watchful Devilkin turned her head, bent
her corselet slantwise, harpooned the Fly, and gripped it
between her two saws. No Cat could pounce more
quickly on a Mouse.
To my surprise I found that the Fly was not only
enough for a meal, but enough for the whole day, and
often for several days. These fierce-looking insects are
extremely abstemious. I was expecting them to be ogres,
and found them with the delicate appetites of invalids.
After a time even a Midge failed to tempt them, and
through the winter months they fasted altogether.
When the spring came, however, they were ready to in-
dulge in a small piece of Cabbage Butterfly or Locust;
attacking their prey invariably in the neck, like the
Mantis.
The young Empusa has one very curious habit when in
captivity. In its cage of wire-gauze its attitude is the
[124]
TWO STRANGE GRASSHOPPERS
same from first to last, and a most strange attitude it is.
It grips the wire by the claws of its four hind-legs, and
hangs motionless, back downwards, with the whole of its
body suspended from those four points. If it wishes to
move, its harpoons open in front, stretch out, grasp a
mesh of the wire, and pull. This process naturally
draws the insect along the wire, still upside down.
Then the jaws close back against the chest.
And this upside-down position, which seems to us so
trying, lasts for no short while. It continues, in my
cages, for ten months without a break. The Fly on the
ceiling, it is true, adopts the same position; but she has
her moments of rest. She flies, she walks in the usual
way, she spreads herself flat in the sun. The Empusa,
on the other hand, remains in her curious attitude for ten
months on end, without a pause. Hanging from the
wire netting, back downwards, she hunts, eats, digests,
dozes, gets through all the experiences of an insect's life,
and finally dies. She clambers up while she is still quite
young; she falls down in her old age, a corpse.
This custom is all the more remarkable in that it is
practised only in captivity. It is not an instmctive
habit of the ra.ce; for out of doors the insect, except at
rare intervals, stands on the bushes back upwards.
Strange as the performance is, I know of a similar case
that is even more peculiar: the attitude of certain
[125]
FABRE'S BOOK OF INSECTS
Wasps and Bees during the night's rest. A particular
Wasp, an Ammophila with red fore-legs, is plentiful in
my enclosure towards the end of August, and likes to
sleep in one of the lavender borders. At dusk, espe-
cially after a stifling day when a storm is brewing, I am
sure to find the strange sleeper settled there. Never was
a more eccentric attitude chosen for a night's rest. The
jaws bite right into the lavender-stem. Its square shape
supplies a firmer hold than a round stalk would give.
With this one and only prop the Wasp's body juts out
stiffly at full length, with legs folded. It forms a right
angle with the stalk, so that the whole weight of the in-
sect rests upon the -mandibles.
The Ammophila is enabled by its mighty jaws to sleep
in this way, extended in space. It takes an animal to
think of a thing like that, which upsets all our previous
ideas of rest. Should the threatening storm burst and
the stalk sway in the wind, the sleeper is not troubled by
her swinging hammock; at most, she presses her fore-legs
for a moment against the tossing stem. Perhaps the
Wasp's jaws, like the Bird's toes, possess the power of
gripping more tightly in proportion to the violence of the
wind. However that may be, there are several kinds of
Wasps and Bees who adopt this strange position, — grip-
ping a stalk with their mandibles, and sleeping with their
bodies outstreched and their legs folded back. This
[126]
TWO STRANGE GRASSHOPPERS
state of things makes us wonder what it is that really con-
stitutes rest.
About the middle of May the Empusa is transformed
into her full-grown condition. She is even more re-
markable in figure and attire than the Praying Mantis.
She still keeps some of her youthful eccentricities — the
bust, the weapons on her knees, and the three rows of
scales on the lower surface of her body. But she is now
no longer twisted into a crook, and is comelier to look
upon. Large pale-green wings, pink at the shoulder and
swift in flight, cover the white and green stripes that orn-
ament the body below. The male Empusa, who is a
dandy, adorns himself, like some of the Moths, with
feathery antennae.
When, in the spring, the peasant meets the Empusa,
he thmks he sees the common Praying Mantis, who is a
daughter of the autumn. They are so much alike that
one would expect them to have the same habits. In fact,
any one might be tempted, led away by the extraordinary
armour, to suspect the Empusa of a mode of life even
more atrocious than that of the Mantis. This would
be a mistake: for all their war-like aspect the Empusae
are peaceful creatures.
Imprisoned in their wire-gauze bell-jar, either m
groups of half a dozen or in separate couples, they at no
time lose their placidity. Even in their full-grown state
[127]
FABRE'S BOOK OF INSECTS
they are very small eaters, and content themselves with
a fly or two as their daily ration.
Big eaters are naturally quarrelsome. The Mantis,
gorged with Locusts, soon becomes irritated and shows
fight. The Ernpusa, with her frugal meals, is a lover of
peace. She indulges in no quarrels with her neighbours,
nor does she pretend to be a ghost, with a view to
frightening them, after the manner of the Mantis. She
never unfurls her wings suddenly nor puffs like a
startled Adder. She has never the least inclination for
the cannibal banquets at which a sister, after being
worsted in a fight, is eaten up. Nor does she, like the
Mantis, devour her husband. Such atrocities are here
unknown.
The organs of the two insects are the same. These
profound moral differences, therefore, are not due to
any difference in the bodily form. Possibly they may
arise from the difference in food. Simple living, as a
matter of fact, softens character, in animals as in men ;
over-feeding brutalises it. The glutton, gorged with
meat and strong drink — a very common cause of savage
outbursts — could never be as gentle as the self-denying
hermit who lives on bread dipped into a cup of milk.
The Mantis is a glutton : the Empusa lives the simple
life.
And yet, even when this is granted, one is forced to
[128]
TWO STRANGE GRASSHOPPERS
ask a further question. Why, when the two insects are
almost exactly the same in form, and might be expected
to have the same needs, should the one have an enormous
appetite and the other such temperate ways? They
tell us, in their own fashion, what many insects have
told us already : that inclinations and habits do not de-
pend entirely upon anatomy. High above the laws that
govern matter rise other laws that govern instincts.
II
THE WHITE-FACED DECTICUS
The White-faced Decticus stands at the head of the
Grasshopper clan in my district, both as a singer and as
an insect of imposing presence. He has a grey body,
a pair of powerful mandibles, and a broad ivory face.
Without being plentiful, he is neither difficult nor weari-
some to hunt. In the height of summer we find him
hopping in the long grass, especially at the foot of the
sunny rocks where the turpentine-tree takes root.
The Greek word dectikos means biting, fond of biting.
The Decticus is well named. It is eminently an insect
given to biting. Mind your finger if this sturdy Grass-
hopper gets hold of it : he will rip it till the blood comes.
His powerful jaw, of which I have to beware when I
handle him, and the large muscles that swell out his
[129]
FABRE'S BOOK OF INSECTS
cheeks, are evidently intended for cutting up leathery
prey.
I find, when the Decticus is imprisoned in my
menagerie, that any fresh meat tasting of Locust or
Grasshopper suits his needs. The blue-winged Locust
is the most frequent victim. As soon as the food is
introduced into the cage there is an uproar, especially
if the Dectici are hungry. They stamp about, and dart
forward clumsily, being hampered by their long shanks.
Some of the Locusts are caught at once, but others with
desperate bounds rush to the top of the cage, and there
hang on out of the reach of the Grasshopper, who is too
stout to climb so high. But they have only postponed
their fate. Either because they are tired, or because
they are tempted by the green stuff below, they will
come down, and the Dectici will be after them im-
mediately.
This Grasshopper, though his intellect is dull, pos-
sesses the art of scientific killing of which we have seen
instances elsewhere. He always spears his prey in the
neck, and, to make it helpless as quickly as possible,
begins by biting the nerves that enable it to move. It
is a very wise method, for the Locust is hard to kill.
Even when beheaded he goes on hopping. I have seen
some who, though half-eaten, kicked out so desperately
that they succeeded in escaping.
[130]
TWO STRANGE GRASSHOPPERS
With his weakness for Locusts, and also for certain
seeds that are harmful to unripe corn, these Grasshoppers
might be of some service to agriculture if only there were
more of them. But nowadays his assistance in preserv-
ing the fruits of the earth is very feeble. His chief
interest in our eyes is the fact that he is a memorial of
the remotest times. He gives us a vague glimpse of
habits now out of use.
It was thanks to the Decticus that I first learnt one
or two things about young Grasshoppers.
Instead of packing their eggs in casks of hardened
foam, like the Locust and the Mantis, or laying them
in a twig like the Cicada, Grasshoppers plant them like
seeds in the earth.
The mother Decticus has a tool at the end of her
body with which she scrapes out a little hole in the soil.
In this hole she lays a certain number of eggs, then
loosens the dust round the side of the hole and rams it
down with her tool, very much as we should pack the
earth in a hole with a stick. In this way she covers up
the well, and then sweeps and smooths the ground above
it.
She then goes for a little walk in the neighbourhood,
by way of recreation. Soon she comes back to the place
where she has already laid her eggs, and, very near the
original spot, which she recognises quite well, begms the
[131]
FABRE'S BOOK OF INSECTS
work afresh. If I watch her for an hour I see her go
through this whole performance, including the short
stroll in the neighbourhood, no less than five times. The
points where she lays the eggs are always very close
together.
When everything is finished I examine the little pits.
The eggs lie singly, without any cell or sheath to protect
them. There are about sixty of them altogether, pale
lilac-grey in colour, and shaped like a shuttle.
When I began to observe the ways of the Decticus
I was anxious to watch the hatching, so at the end of
August I gathered plenty of eggs, and placed them in
a small glass jar with a layer of sand. Without suffer-
ing any apparent change they spent eight months there
under cover, sheltered from the frosts, the showers, and
the overpowering heat of the sun, which they would be
obliged to endure out of doors.
When June came, the eggs in my jar showed no sign
of being about to hatch. They were just as I had
gathered them nine months before, neither wrinkled nor
tarnished, but on the contrary wearing a most healthy
look. Yet in June young Dectici are often to be met
in the fields, and sometimes even those of larger growth.
What was the reason of this delay, I wondered.
Then an idea came to me. The eggs of the Grass-
hopper are planted like seeds in the earth, were they are
[132]
TWO STRANGE GRASSHOPPERS
exposed, without any protection, to snow and rain.
Those in my jar had spent two-thirds of the year in a
state of comparative dryness. Since they were sown
like seeds, perhaps they needed, to make them hatch,
the moisture that seeds require to make them sprout.
I resolved to try. ^
I placed at the bottom of some glass tubes a pinch
of backward eggs taken from my collection, and on the
top I heaped lightly a layer of fine, damp sand. I closed
the tubes with plugs of wet cotton, to keep the air in them
constantly moist. Any one seeing my preparations
would have supposed me to be a botanist experimenting
with seeds.
My hopes were fulfilled. In the warmth and mois-
ture the eggs soon showed signs of hatching. They
began to swell, and the bursting of the shell was evi-
dently close at hand. I spent a fortnight in keeping
a tedious watc^h at every hour of the day, for I had to
surprise the young Decticus actually leaving the egg,
in order to solve a question that had long been in my
mind.
The question was this. The Grasshopper is buried,
as a rule, about an inch below the surface of the soil.
Now the new-born Decticus, hopping awkwardly in the
grass at the approach of summer, has, like the full-grown
insect, a pair of very long tentacles, as slender as hairs:
[133]
FABRE'S BOOK OF INSECTS
while he carries behind him two extraordinary legs, two
enormous hinged jumping-poles that would be very in-
convenient for ordinary walking. I wished to find out
how the feeble little creature set to work, with this
cumbrous luggage, to make its way to the surface of the
earth. By what means could it clear a passage through
the rough soil^ With its feathery antennae, which an
atom of sand can break, and its immense shanks, which
are disjointed by the least effort, this mite is plainly
incapable of freeing itself.
As I have already told you, the Cicada and the
Praying Mantis, when issuing, the one from his twig,
and the other from his nest, wear a protective covering
like an overall. It seemed to me that the little Grass-
hopper, too, must come out through the sand in a simpler,
more compact form than he wears when he hops about the
lawn on the day after his birth.
Nor was I mistaken. The Decticus, like the others,
wears an overall for the occasion. The tiny, flesh-white
creature is cased in a scabbard which keeps the six legs
flattened against the body, stretching backwards, inert.
In order to slip more easily through the soil his shanks
are tied up beside him; while the antennae, those other
inconvenient appendages, are pressed motionless against
the parcel.
The head is very much bent against the chest. With
[134]
TWO STRANGE GRASSHOPPERS
the big black specks that are going to be its eyes, and its
inexpressive, rather swollen mask, it suggests a diver's
helmet. The neck opens wide at the back, and, with a
slow throbbing, by turns swells and sinks. It is by
means of this throbbing protrusion through the opening at
the back of the head that the new-born insect moves.
When the lump is flat, the head pushes back the damp
sand a little way and slips into it by digging a tiny pit.
Then the swelling is blown out and becomes a knob
which sticks firmly in the hole. This supplies the re-
sistance necessary for the grub to draw up its back and
push. Thus a step forward is made. Each thrust of
the motor-blister helps the little Decticus upon the up-
ward path.
It is pitiful to see this tender creature, still almost
colourless, knocking with its swollen neck and ramming
the rough soil. With flesh that is not yet hardened it
is painfully fighting stone; and fighting it so success-
fully that in the space of a morning it makes a gallery,
either straight or winding, an inch long and as wide as
an average straw. In this way the harassed insect
reaches the surface.
Before it is altogether freed from the soil the struggler
halts for a moment, to recover from the effects of the
journey. Then, with renewed strength, it makes a last
effort: it swells the protrusion at the back of its head as
[135]
FABRE'S BOOK OF INSECTS
far as it will go, and bursts the sheath that has protected
it so far. The creature throws off its overall.
Here, then, is the Decticus in his youthful shape,
quite pale still, but darker the next day, and a regular
blackamoor compared with the full-grown insect. As a
prelude to the ivory face of his riper age he wears a
narrow white stripe under his hinder thighs.
Little Decticus, hatched before my eyes, life opens
for you very harshly I Many of your relatives must die
of exhaustion before winning their freedom. In my
tubes I see numbers who, being stopped by a grain of
sand, give up the struggle half-way and become furred
with a sort of silky fluff. Mildew soon absorbs their
poor little remains. And when carried out without my
help, their journey to the surface must be even more
dangerous, for the soil out of doors is coarse and baked
by the sun.
The little white-striped nigger nibbles at the lettuce-
leaf I give him, and leaps about gaily in the cage where
I have housed him. I could easily rear him, but he
would not teach me much more. So I restore him to
liberty. In return for what he has taught me I give
him the grass and the Locusts in the garden.
For he taught me that Grasshoppers, in order to
leave the ground where the eggs are laid, wear a tem-
porary form which keeps those too cumbrous parts, the
[136]
TWO STRANGE GRASSHOPPERS
long legs and antennae, swathed together in a sheath.
He taught me, too, that this mummy-like creature, fit
only to lengthen and shorten itself a little, has for its
means of travelling a hernia in the neck, a throbbing
blister — an original piece of mechanism which, when I
first observed the Decticus, I had never seen used as an
aid to progression.
[137]
CHAPTER X
COMMON WASPS
THEIR CLEVERNESS AND STUPIDITY
WISHING to observe a Wasp's nest I go
out, one day in September, with my
little son Paul, who helps me with his
good sight and his undivided attention. We look with
interest at the edges of the footpaths.
Suddenly Paul cries: "A Wasp's nest I A Wasp's
nest, as sure as anything I" For, twenty yards away,
he has seen rising from the ground, shooting up and
flying away, now one and then another swiftly moving
object, as though some tiny crater in the grass were
hurling them forth.
We approach the spot with caution, fearing to attract
the attention of the fierce creatures. At the entrance-
door of their dwelling, a round opening large enough to
admit a man's thumb, the inmates come and go, busily
passing one another as they fly in opposite directions.
Burr! A shudder runs through me at the thought of the
unpleasant time we should have, did we incite these
[138]
COMMON WASPS
irritable warriors to attack us by inspecting them too
closely. Without further investigation, which might
cost us too dear, we mark the spot, and resolve to return
at nightfall. By that time all the inhabitants of the
nest will have come home from the fields.
The conquest of a nest of Common Wasps would be
rather a serious undertaking if one did not act with a
certain amount of prudence. Half a pint of petrol,
a reed-stump nine inches long, and a good-sized lump
of clay or loam, kneaded to the right consistency — such
are my weapons, which I have come to consider the best
and simplest, after various trials with less successful
means.
The suffocating method is necessary, unless I use costly
measures which I cannot afford. When Reaumur
wanted to place a live Wasp's nest in a glass case with a
view to observing the habits of the inmates, he employed
helpers who were used to the painful job, and were
willing, for a handsome reward, to serve the man of
science at the cost of their skins. But I, who should
have to pay with my own skin, think twice before dig-
ging up the nest I desire. I begin by suffocating the
inhabitants. Dead Wasps do not sting. It is a brutal
method, but perfectly safe.
I use petrol because its effects are not too violent, and
in order to make my observations I wish to leave a small
[139]
FABRE'S BOOK OF INSECTS
numjer of survivors. The question is how to intro-
duce it into the cavity containing the Wasp's nest. A
vestibule, or entrance-passage, about nine inches long,
and very nearly horizontal, leads to the underground
cells. To pour the petrol straight into the mouths of
this tunnel would be a blunder that might have serious
consequences later on. For so small a quantity of petrol
would be absorbed by the soil and would never reach
the nest; and next day, when we might think we were
digging safely, we should find an infuriated swarm
under the spade.
The bit of reed prevents this mishap. When in-
serted into the passage it forms a water-tight funnel,
and carries the petrol to the cavern without the loss of
a drop, and as quickly as possible. Then we fix the
lump of kneaded clay into the entrance-hole, like a
stopper. We have nothing to do now but wait.
When we are going to perform this operation Paul
and I set out, carrying a lantern and a basket with the
implements, at nine o'clock on some mild, moonlit
evening. While the farm-house Dogs are yelping at
each other in the distance, and the Screech Owl is hooting
in the olive-trees, and the Italian Crickets are performing
their symphony in the bushes, Paul and I chat about
insects. He asks questions, eager to learn, and I tell
[140]
COMMON WASPS
him the little that I know. So delightful are our nights
of Wasp-hunting that we think little of the loss of sleep
or the chance of being stung I
The pushing of the reed into the hole is the most
delicate matter. Since the direction of the passage is
unknown there is some hesitation, and sometimes sentries
come flying out of the Wasp's guard-house to attack the
operator's hand. To prevent this one of us keeps watch,
and drives away the enemy with a handkerchief. And
after all, a swelling on one's hand, even if it does smart,
is not much to pay for an idea.
As the petrol streams into the cavern we hear the
threatening buzz of the population underground. Then
quick ! — the door must be closed with the wet clay, and
the clod kicked once or twice with the heel to make the
stopper solid. There is nothing more to be done for
the present. Off we go to bed.
With a spade and a trowel we are back on the spot at
dawn. It is wise to be early, because many Wasps will
have been out all night, and will want to get into their
home while we are digging. The chill of the morning
will make them less fierce.
In front of the entrance-passage, in which the reed
is still sticking, we dig a trench wide enough to allow us
free movement. Then the side of this ditch is carefully
[141]
FABRE'S BOOK OF INSECTS
cut away, slice after slice, until, at a depth of about
twenty inches, the Wasp's nest is revealed, uninjured,
slung from the roof of a spacious cavity.
It is indeed a superb achievement, as large as a fair-
sized pumpkin. It hangs free on every side except at
the top, where various roots, mostly of couch-grass, pene-
trate the thickness of the wall and fasten the nest firmly.
Its shape is round wherever the ground has been soft,
and of the same consistency all through. In stony soil,
where the Wasps meet with obstacles in their digging,
the sphere becomes more or less misshapen.
A space of a hand's-breadth is always left open be-
tween the paper nest and the sides of the underground
vault. This space is the wide street along which the
builders move unhindered at their continual task of
enlarging and strengthening the nest, and the passage
that leads to the outer world opens into it. Under-
neath the nest is a much larger unoccupied space,
rounded into a big basin, so that the wrapper of the nest
can be enlarged as fresh cells are added. This cavity
also serves as a dust-bin for refuse.
The cavity was dug by the Wasps themselves. Of
that there is no doubt; for holes so large and so regular
do not exist ready-made. The original foundress of the
nest may have seized on some cavity made by a Mole,
to help her at the beginning ; but the greater part of the
[142]
COMMON WASPS
enormous vault was the work of the Wasps. Yet there
is not a scrap of rubbish outside the entrance. Where
is the mass of earth that has been removed'?
It has been spread over such a large surface of ground
that it is unnoticed. Thousands and thousands of
Wasps work at digging the cellar, and enlarging it as
that becomes necessary. They fly up to the outer world,
each carrying a particle of earth, which they drop on the
ground at some distance from the nest, in all directions.
Being scattered in this way the earth leaves no visible
trace.
The Wasp's nest is made of a thin, flexible material
like brown paper, formed of particles of wood. It is
streaked with bands, of which the colour varies according
to the wood used. If it were made in a single continu-
ous sheet it would give little protection against the cold.
But the Common Wasp, like the ballon-maker, knows
that heat may be preserved by means of a cushion of
air contained by several wrappers. So she makes her
paper-pulp into broad scales, which overlap loosely and
are laid on in numerous layers. The whole forms a
coarse blanket, thick and spongy in texture and well
filled with stagnant air. The temperature under this
shelter must be truly tropical in hot weather.
The fierce Hornet, chief of the Wasps, builds her nest
on the same principle. In the hollow of a willow, or
[143]
FABRE'S BOOK OF INSECTS
within some empty granary, she makes, out of fragments
of wood, a very brittle kind of striped yellow cardboard.
Her nest is wrapped round with many layers of this
substance, laid on in the form of broad convex scales
which are welded to one another. Between them are
wide intervals in which air is held motionless.
The Wasp, then, often acts in accordance with the
laws of physics and geometry. She employs air, a non-
conductor of heat, to keep her home warm; she made
blankets before man thought of it ; she builds the outer
walls of the nest in the shape that gives her the largest
amount 'of room in the smallest wrapper; and in the
form of her cell, too, she economises space and material.
And yet, clever as these wonderful architects are, they
amaze us by their stupidity in the face of the smallest
difficulty. On the one hand their instincts teach them
to behave like men of science; but on the other it is plain
that they are entirely without the power of reflection.
I have convinced myself of this fact by various experi-
ments.
The Common Wasp has chanced to set up house be-
side one of the walks in my enclosure, which enables
me to experiment with a bell-glass. In the open fields
I could not use this appliance, because the boys of the
country-side would soon smash it. One night, when all
was dark and the Wasps had gone home, I placed the
[144]
COMMON WASPS
glass over the entrance of the burrow, after first flat-
tening the soil. When the Wasps began work again
next morning and found themselves checked in their
flight, would they succeed in making a passage under
the rim of the glass? Would these sturdy creatures,
who were capable of digging a spacious cavern, reahse
that a very short underground tunnel would set them
free? That was the question.
The next morning I found the bright sunlight falling
on the bell-glass, and the workers ascending in crowds
from underground, eager to go in search of provisions.
They butted against the transparent wall, tumbled
down, picked themselves up again, and whirled round
and round in a crazy swarm. Some, weary of dancing,
wandered peevishly at random and then re-entered their
dwelling. Others took their places as the sun grew
hotter. But not one of them, not a single one, scratched
with her feet at the base of the glass circle. This means
of escape was beyond them.
Meanwhile a few Wasps who had spent the night
out of doors were coming in from the fields. Round and
round the bell-glass they flew; and at last, after much
hesitation, one of them decided to dig under the edge.
Others followed her example, a passage was easily
opened, and the Wasps went in. Then I closed the
passage with some earth. The narrow opening, if seen
[145]
FABRE'S BOOK OF INSECTS
from within, might help the Wasps to escape, and I
wished to leave the prisoners the honour of winning
their liberty.
However poor the Wasps' power of reasoning, I
thought their escape was now probable. Those who had
just entered would surely show the way; they would
teach the others to dig below the wall of glass.
I was too hasty. Of learning by experience or ex-
ample there was not a sign. Inside the glass not an
attempt was made to dig a tunnel. The insect popula-
tion whirled round and round, but showed no enter-
prise. They floundered about, while every day numbers
died from famine and heat. At- the end of a week not
one was left alive. A heap of corpses covered the
ground.
The Wasps returning from the field could find their
way in, because the power of scenting their house
through the soil, and searching for it, is one of their
natural instincts, one of the means of defence -given to
them. There is no need for thought or reasoning here :
the earthy obstacle has been familiar to every Wasp
since Wasps first came into the world.
But those who are within the bell-glass have no such
instinct to help them. Their aim is to get into the light,
and finding daylight in their transparent prison they
think their aim is accomplished. In spite of constant
[146]
COMMON WASPS
collisions with the glass they spend themselves in vainly
trying to fly farther in the direction of the sunshine.
There is nothing in the past to teach them what to do.
They keep blindly to their familiar habits, and die.
SOME OF THEIR HABITS
If we open the thick envelope of the nest we shall find,
inside, a number of combs, or layers of cells, lying one
belo-w the other and fastened together by solid pillars.
The number of these layers varies. Towards the end of
the season there may be ten, or even more. The opening
of the cells is on the lower surface. In this strange
world the young grow, sleep, and receive their food
head downwards.
The various storeys, or layers of combs, are divided
by open spaces; and between the outer envelope and
the stack of combs there are doorways through which
every part can be easily reached. There is a continual
coming and going of nurses, attending to the grubs in
the cells. On one side of the outer wrapper is the gate
of the city, a modest unadorned opening, lost among
the thin scales of the envelope. Facing it is the entrance
to the tunnel that leads from the cavity to the world
at large.
r [147]
FABRE'S BOOK OF INSECTS
In a Wasp community there is a large number of
Wasps whose whole life is spent in work. It is their
business to enlarge the nest as the population grows;
and though they have no grubs of their own, they nurse
the grubs in the cells with the greatest care and industry.
Wishing to watch their operations, and also to see what
would take place at the approach of winter, I placed
under cover one October a few fragments of a nest, con-
taking a large number of eggs and grubs, with about
a hundred workers to take care of them.
To make my inspection easier I separated the combs
and placed them side by side, with the openings of the
cells turned upwards. This arrangement, the reverse
of the usual position, did not seem to annoy my
prisoners, who soon recovered from the disturbance and
set to work as if nothing had happened. In case they
should wish to build I gave them a slip of soft wood;
and I fed them with honey. The underground cave in
which the nest hangs out of doors was represented by
a large earthen pan under a wire-gauze cover. A re-
movable cardboard dome provided darkness for the
Wasps, and — when removed — light for me.
The Wasps' work went on as if it had never been
interrupted. The worker- Wasps attended to the grubs
and the building at the same time. They began to
raise a wall round the most thickly populated combs;
[148]
COMMON WASPS
and it seemed as though they might intend to build a
new envelope, to replace the one ruined by my spade.
But they were not repairing; they were simply carrying
on the work from the point at which I interrupted it.
Over about a third of the comb they made an arched roof
of paper scales, which would have been joined to the
envelope of the nest if it had been intact. The tent
they made sheltered only a small part of the disk of
cells.
As for the wood I provided for them, they did not
touch it. To this raw material, which would have been
troublesome to work, they preferred the old cells that
were no longer in use. In these the fibres were already
prepared; and, with a little saliva and a little grinding
in their mandibles, they turned them into pulp of the
highest quality. The uninhabited cells were nibbled
into pieces, and out of the ruins a sort of canopy was
built. New cells could be made in the same way if
necessary.
Even more interesting than this roofing-work is the
feeding of the grubs. One could never weary of the
sight of the rough fighters turned into tender nurses.
The barracks become a creche. With what care those
grubs are reared! If we watch one of the busy Wasps
we shall see her, with her crop swollen with honey, halt
in front of a cell. With a thoughtful air she bends
[149]
FABRE'S BOOK OF INSECTS
her head into the opening, and touches the grub with the
tip of her antenna. The grub wakes and gapes at her,
like a fledgling when the mother-bird returns to the
nest with food.
For a moment the awakened larva swings its head to
and fro: it is blind, and is trying to feel the food brought
to it. The two mouths meet; a drop of syrup passes
from the nurse's mouth to the nurseling's. That is
enough for the moment: now for the next Wasp-baby.
The nurse moves on, to continue her duties elsewhere.
Meanwhile the grub is licking the base of its own
neck. For, while it is being fed, there appears a tempo-
rary swelling on its chest, which acts as a bib, and catches
whatever trickles down from the mouth. After
swallowing the chief part of the meal the grub gathers up
the crumbs that have fallen on its bib. Then the swell-
ing disappears; and the grub, withdrawing a little way
into its cell, resumes its sweet slumbers.
When fed in my cage the Wasp-grubs have their heads
up, and what falls from their mouths collects naturally
on their bibs. When fed in the nest they have their
heads down. But I have no doubt that even in this
position the bib serves its purpose.
By slightly bending its head the grub can always de-
posit on the projecting bib a portion of the overflowing
mouthful, which is sticky enough to remain there.
[150]
COMMON WASPS
Moreover, it is quite possible that the nurse herself
places a portion of her helping on this spot. Whether
it be above or below the mouth, right way up or upside
down, the bib fulfils its office because of the sticky nature
of the food. It is a temporary saucer which shortens the
work of serving out the rations, and enables the grub to
feed in a more or less leisurely fashion and without too
much gluttony.
In the open country, late in the year when fruit is
scarce, the grubs are mostly fed upon minced Fly; but in
my cages everything is refused but honey. Both nurses
and nurselings seem to thrive on this diet, and if any in-
truder ventures too near to the combs he is doomed.
Wasps, it appears, are far from hospitable. Even the
Polistes, an insect who is absolutely like a Wasp in shape
and colour, is at once recognised and mobbed if she
approaches the honey the Wasps are sipping. Her ap-
pearance takes nobody in for a moment, and unless she
hastily retires she will meet with a violent death. No,
it is not a good thing to enter a Wasps' nest, even when
the stranger wears the same uniform, pursues the same
industry, and is almost a member of the same corpo-
ration.
Again and again I have seen the savage reception given
to strangers. If the stranger be of sufficient importance
he is stabbed, and his body is dragged from the nest and
[151]
FABRE'S BOOK OF INSECTS
flung into the refuse-heap below. But the poisoned dag-
ger seems to be reserved for great occasions. If I throw
the grub of a Saw-fly among the Wasps they show great
surprise at the black-and-green dragon; they snap at it
boldly, and wound it, but without stinging it. They try
to haul it away. The dragon resists, anchoring itself to
the comb by its hooks, holding on now by its fore-legs
and now by its hind-legs. At last the grub, however,
weakened by its wounds, is torn from the comb and
dragged bleeding to the refuse-pit. It has taken a
couple of hours to dislodge it.
Supposing, on the other hand, I throw on to the combs
a certain imposing grub that lives under the bark of
cherry-trees, five or six Wasps will at once prick it with
their stings. In a couple of minutes it is dead. But the
hugh dead body is much too heavy to be carried out of
the nest. So the Wasps, finding they cannot move
the grub, eat it where it lies, or at least reduce its weight
till they can drag the remains outside the walls.
Ill
THEIR SAD END
Protected in this fierce way against the invasion of in-
truders, and fed with excellent honey, the grubs in my
cage prosper greatly. But of course there are excep-
[152]
COMMON WASPS
tions. In the Wasps' nest, as everywhere, there are
weaklings who are cut down before their time.
I see these puny sufferers refuse their food and slowly
pine away. The nurses perceive it even more clearly.
They -bend their heads over the invalid, sound it with
their antennae, and pronounce it incurable. Then the
creature at the point of death is torn ruthlessly from its
cell and dragged outside the nest. In the brutal com-
monwealth of the Wasps the invalid is merely a piece
of rubbish, to be got rid of as soon as possible for fear of
contagion. Nor indeed is this the worst. As winter
draws near the Wasps foresee their fate. They know
their end is at hand.
The first cold nights of November bring a change in
the nest. The building proceeds with diminished en-
thusiasm; the visits to the pool of honey are less constant.
Household duties are relaxed. Grubs gaping with
hunger receive tardy relief, or are even neglected. Pro-
found uneasiness seizes upon the nurses. Their former
devotion is succeeded by indifference, which soon turns
to dislike. What is the good of continuing attentions
which soon will be impossible? A time of famine is
coming; the nurselings in any case must die a tragic
death. So the tender nurses become savage execu-
tioners.
-Let us leave no orphans," they say to themselves;
[153]
FABRE'S BOOK OF INSECTS
'no one would care for them after we are gone. Let us
kill everything, eggs and grubs alike. A violent end is
better than a slow death by starvation."
A massacre follows. The grubs are seized by the
scruff of the neck, brutally torn from their cells, dragged
out of the nest, and thrown into the refuse-heap at the
bottom of the cave. The nurses, or workers, root them
out of their cells as violently as though they were
strangers or dead bodies. They tug at them savagely
and tear them. Then the eggs are ripped open and de-
voured.
Before much longer the nurses themselves, the execu-
tioners, are languidly dragging what remains of their
lives. Day by day, with a curiosity mingled with
emotion, I watch the end of my insects. The workers
die suddenly. They come to the surface, slip down, fall
on their backs and rise no more, as if they were struck
by lightning. They have had their day; they are slain
by age, that merciless poison. Even so does a piece of
clockwork become motionless when its mainspring has
unwound its last spiral.
The workers are old : but the mothers are the last to
be born into the nest, and have all the vigour of youth.
And so, when winter sickness seizes them, they are
capable of a certain resistance. Those whose end is near
are easily distinguished from the others by the disorder
[154]
COMMON WASPS
of their appearance. Their backs are dusty. While
they are well they dust themselves without ceasing, and
their black-and-yellow coats are kept perfectly glossy.
Those who are ailing are careless of cleanliness; they
stand motionless in the sun or wander languidly about.
They no longer brush their clothes.
This indifference to dress is a bad sign. Two or three
days later the dusty female leaves the nest for the last
time. She goes outside, to enjoy yet a little of the sun-
light; presently she slides quietly to the ground and
does not get up again. She declines to die in her be-
loved paper home, where the code of the Wasps ordains
absolute cleanliness. The dying Wasp performs her
own funeral rites by dropping herself into the pit at
the bottom of the cavern. For reasons of health these
stoics refuse to die in the actual house, among the combs.
The last survivors retain this repugnance to the very
end. It is a law that never falls into disuse, however
greatly reduced the population may be.
My cage becomes emptier day by day, notwithstand-
ing the mildness of the room, and notwithstanding the
saucer of honey at which the able-bodied come to sip.
At Christmas I have only a dozen fenrtales left. On the
sixth of January the last of them perishes.
Whence arises this mortality, which mows down the
whole of my wasps'? They have not suffered from
[155]
FABRE'S BOOK OF INSECTS
famine : they have not suffered from cold : they have not
suffered from home-sickness. Then what have they
died of?
We must not blame their captivity. The same thing
happens in the open country. Various nests I have in-
spected at the end of December all show the same condi-
tion. The vast majority of Wasps must die, apparently,
not by accident, nor illness, nor the inclemency of the
season, but by an inevitable destiny, which destroys them
as energetically as it brings them into life. And it is
well for us that it is so. One female Wasp is enough
to found a city of thirty thousand inhabitants. If all
were to survive, what a scourge they would be I The
Wasps would tyrannise over the countryside.
In the end the nest itself perishes. A certain Cater-
pillar which later on becomes a mean-looking Moth;
a tiny reddish Beetle; and a scaly grub clad in gold
velvet, are the creatures that demolish it. They gnaw
the floors of the various storeys, and crumble the whole
dwelling. A few pinches of dust, a few shreds of brown
paper are all that remain, by the return of spring, of the
Wasps' city and its thirty thousand inhabitants.
[156]
CHAPTER XI
THE AVENTURES OF A GRUB
THE YOUNG SITARIS
THE high banks of sandy clay in the country
round about Carpentras are the favourite
haunts of a host of Bees and Wasps, those
lovers of a sunny aspect and of soil that is easy to dig
in. Here, in the month of May, two Bees, both of them
Mason-bees, builders of subterranean cells, are espe-
cially abundant. One of them builds at the entrance of
her dwelling an advanced fortification, an earthly
cylinder, wrought in open work and curved, of the
width and length of a man's finger. When it is peopled
with many Bees one stands amazed at the elaborate
ornamentation formed by all these hanging fingers of
clay.
The other Bee, who is very much more frequently
seen and is called Anthophora pilipes, leaves the opening
of her corridor bare. The chinks between the stones
in old walls and abondoned hovels, or exposed surfaces
of sand stone or marl, are found suitable for her labours;
[157]
FABRE'S BOOK OF INSECTS
but the favourite spots, those to which the greatest
number of swarms resort, are straight stretches of ground
exposed to the south, such as occur in the cuttings of
deeply-sunken roads. Here, over areas many yards in
width, the wall is drilled with a multitude of holes,
which give to the earthy mass the look of some enormous
sponge. These round holes might have been made with
a gimlet, so regular are they. Each is the entrance to
a winding corridor, which runs to the depth of four or
hve inches. The cells are at the far end. If we wish
to watch the labours of the industrious Bee we must
visit her workshop during the latter half of May. Then
— but at a respectful distance — we may see, in all its
bewildering activity, the tumultuous, buzzing swarm,
busied with the building and provisioning of the cells.
But it has been most often during the months of
August and September, the happy months of the summer
holidays, that I have visited the banks inhabited by the
Anthophora. At this season all is silent near the nests :
the work has long been completed: and numbers of
Spiders' webs line the crevices or plunge their silken
tubes into the Bees' corridors. That is no reason, how-
ever, for hastily abandoning the city that was once so
full of life and bustle, and now appears deserted. A
few inches below the surface, thousands of grubs are
imprisoned in their cells of clay, resting until the coming
[158]
THE ADVENTURES OF A GRUB
spring. Surely these grubs, which are paralysed and
incapable of self-defence, must be a temptation — fat
little morsels as they are — to some kind of parasite, some
kind of insect stranger in search of prey. The matter
is worth inquiring into.
Two facts are at once noticeable. Some dismal-
looking Flies, half black and half white, are flying in-
dolently from gallery to gallery, evidently with the
object of laying their eggs there. Many of them are
hanging dry and lifeless in the Spiders' webs. At other
places the entire surface of a bank is hung with the dried
corpses of a certain Beetle, called the Sitaris. Among
the corpses, however, are a few live Beetles, both male
and female. The female Beetle invariably disappears
into the Bees' dwelling. Without a doubt she, too, lays
her eggs there.
If we give a few blows of the pick to the surface of
the bank we shall find out something more about these
things. During the early days of August this is what
we shall see : the cells forming the top layer are unlike
those at a greater depth. The difference is owing to the
fact that the same establishment is used by two kinds of
Bee, the Anthophora and the Osmia.
The Anthophorae are the actual pioneers. The work
of boring the galleries is wholly theirs, and their cells
are right at the end. If they, for any reason, leave the
[159]
FABRE'S BOOK OF INSECTS
outer cells, the Osmia comes in and takes possession of
them. She divides the corridors into unequal and in-
artistic cells by means of rough earthen partitions, her
only idea of masonry.
The cells of the Anthophora are faultlessly regular and
perfectly finished. They *are works of art, cut out of
the very substance of the earth, well out of reach of all
ordinary enemies; and for this reason the larva of this
Bee has no means of spinning a cocoon. It lies naked
in the cell, whose inner surface is polished like stucco.
In the Osmia's cells, however, means of defence are
required, because they are at the surface of the soil, are
roughly made, and are badly protected by their thin par-
titions. So the Osmia's grubs enclose themselves in a
very strong cocoon, which preserves them both from the
rough sides of their shapeless cells and from the jaws
of various enemies who prowl about the galleries. It
is easy, then, in a bank inhabited by these two Bees,
to recognise the cells belonging to each. The An-
thophora's cells contain a naked grub : those of the Osmia
contain a grub enclosed in a cocoon.
Now each of these two Bees has its own especial
parasite, or uninvited guest. The parasite of the Osmia
is the black-and-white Fly who is to be seen so often
at the entrance to the galleries, intent on laying her eggs
within them. The parasite of the Anthophora is the
[160]
THE ADVENTURES OF A GRUB
Sitaris, the Beetle whose corpses appear in such quan-
tities on the surface of the bank.
If the layer of Osmia-cells be removed from the nest
we can observe the cells of the Anthophora. Some will
be occupied by larvae, some by the perfect insect, and
some — indeed many — will contain a singular egg-shaped
shell, divided into segments with projecting breathing-
pores. This shell is extremely thin and fragile; it is
amber-coloured, and so transparent that one can dis-
tinguish quite plainly through its sides a full-grown
Sitaris, struggling as though to set herself at liberty.
What is this curious shell, which does not appear to
be a Beetle's shell at all*? And how can this parasite
reach a cell which seems to be inaccessible because of
its position, and in which the most careful examination
under the magnifying-glass reveals no sign of violence'?
Three years of close observation enabled me to answer
these questions, and to add one of its most astonishing
chapters to the story of insect life. Here is the result
of my inquiries.
The Sitaris in the full-grown state lives only for a day
or two, and its whole life is passed at the entrance to
the Anthophora's galleries. It has no concern but the
reproduction of the species. It is provided with the
usual digestive organs, but I have grave reasons to doubt
whether it actuallv takes any nourishment whatever.
[161]
FABRE'S BOOK OF INSECTS
The female's only thought is to lay her eggs. This done,
she dies. The male, after cowering in a crevice for a day
or two, also perishes. This is the origin of all those
corpses swinging in the Spiders' web, with which the
neighbourhood of the Anthophora's dwelling is uphol-
stered.
At first sight one would expect that the Sitaris, when
laying her eggs, would go from cell to cell, confiding
an egg to each of the Bee-grubs. But when, in the
course of my observations, I searched the Bees' galleries,
I invariably found the eggs of the Sitaris gathered in
a heap inside the entrance, at a distance of an inch or
two from the opening. They are white, oval, and very
small, and they stick together slightly. As for their
number, I do not believe I am exaggerating when I esti-
mate it at two thousand at least.
Thus, contrary to what one was to some extent en-
titled to suppose, the eggs are not laid in the cells of the
Bee; they are simply dumped in a heap inside the door-
way of her dwelling. Nay more, the mother does not
make any protective structure for them; she takes no
pains to shield them from the rigours of winter; she
does not even attempt to stop up the entrance-lobby
in which she has placed them, and so protect them from
the thousand enemies that threaten them. For as long
[162]
THE ADVENTURES OF A GRUB
as the frosts of winter have not arrived these open gal-
leries are trodden by Spiders and other plunderers, for
whom the eggs would make an agreeable meal.
The better to observe them, I placed a number of the
eggs in boxes; and when they hatched out about the
end of September I imagined they would at once start
off in search of an Anthophora-cell. I was entirely
wrong. The young grubs — little black creatures no
more than the twenty-fifth of an inch long — did not
move away, though provided with vigorous legs. They
remained higgledy-piggledy, mixed up with the skins of
the eggs whence they came. In vain I placed within
their reach lumps of earth containing open Bee-cells:
nothing would tempt them to move. If I forcibly re-
moved a few from the common heap they at once hur-
ried back to it in order to hide themselves among the
rest.
At last, to assure myself that the Sitaris-grubs, in the
free state, do not disperse after they are hatched, I went
in the winter to Carpentras and inspected the banks
inhabited by the Anthophorae. There, as in my boxes,
I found the grubs all piled up in heaps, all mixed up
with the skins of the eggs.
I was no nearer answering the question: how does the
Sitaris get into the Bees' cells, and into a shell that does
not belong to it?
[163]
FABRE'S BOOK OF INSECTS
II
THE FIRST ADVENTURE
The appearance of the young Sitaris showed me at
once that its habits must be peculiar. It could not, I
saw, be called on to move on an ordinary surface. The
spot where this larva has to live evidently exposes it
to the risk of many dangerous falls, since, in order to
prevent them, it is equipped with a pair of powerful
mandibles, curved and sharp; robust legs which end in
a long and very mobile claw; a variety of bristles and
probes; and a couple of strong spikes with sharp, hard
points — an elaborate mechanism, like a sort of plough-
share, capable of biting into the most highly polished
surface. Nor is this all. It is further provided with
a sticky liquid, sufficiently adhesive to hold it in position
without the help of other appliances. In vain I racked
my brains to guess what the substance might be, so
shifting, so uncertain, and so perilous, which the young
Sitaris is destined to inhabit. I waited with eager im-
patience for the return of the warm weather.
At the end of April the young grubs imprisoned in my
cages, hitherto lying motionless and hidden in the spongy
heap of egg-skins, suddenly began to move. They
scattered, and ran about in all directions through the
boxes and jars in which they have passed the winter.
[164]
THE ADVENTURES OF A GRUB
Their hurried movements and untiring energy showed
they were in search of something, and the natural
thing for them to seek was food. For these grubs were
hatched at the end of September, and since then, that is
to say for seven long months, they had taken no nourish-
ment, although they were by no means in a state of
torpor. From the moment of their hatching they are
doomed, though full of life, to an absolute fast lasting
for seven months; and when I saw their excitement I
naturally supposed that an imperious hunger had set
them bustling in that fashion.
The food they desired could only be the contents of
the Anthophora's cells, since at a later stage the Sitaris
is found in those cells. Now these contents are limited
to honey and Bee-grubs.
I offered them some cells containing larvae: I even
slipped the Sitares into the cells, and did all sorts of
things to tempt their appetite. My efforts were fruit-
less. Then I tried honey. In hunting for cells pro-
visioned with honey I lost a good part of the month of
May. Having found them I removed the Bee-grub
from some of them, and laid the Sitaris-grub on the
surface of the honey. Never did experiment break
down so completely! Far from eating the honey, the
grubs became entangled in the sticky mass and perished
in it, suffocated. "I have offered you larva, cells,
[165]
FABRE'S BOOK OF INSECTS
honey I" I cried in despair. "Then what do you want,
you fiendish little creatures'?"
Well, in the end I found out what they wanted.
They wanted the Anthophora herself to carry them into
the cells !
When April comes, as I said before, the heap of grubs
at the entrance to the Bees' cells begins to show signs of
activity. A few days later they are no longer there.
Strange as it may appear, they are all careering about
the country, sometimes at a great distance, clinging like
grim death to the fleece of a Bee !
When the Anthophorae pass by the entrance to their
cells, on their way either in or out, the young Sitaris-
grub, who is lying in wait there, attaches himself to
one of the Bees. He wriggles into the fur and clutches
it so firmly that he need not fear a fall during the long
journeys of the insect that carries him. By thus attach-
ing himself to the Bee the Sitaris intends to get himself
carried, at the right moment, into a cell supplied with
honey.
One might at first sight believe that these adventur-
ous grubs derive food for a time from the Bee's body.
But not at all. The young Sitares, embedded in the
fleece, at right angles to the body of the Anthophora,
head inwards, tail outwards, do not stir from the spot
they have selected, a point near the Bee's shoulders.
[166]
THE ADVENTURES OF A GRUB
We do not see them wandering from spot to spot, ex-
ploring the Bee's body, seeking the part where the skin
is most delicate, as they would certainly do if they were
really feeding on the insect. On the contrary, they
are always fixed on the toughest and hardest part of the
Bee's body, a little below the insertion of the wings,
or sometimes on the head; and they remain absolutely
motionless, clinging to a single hair. It seems to me
undeniable that the young Sitares settle on the Bee
merely to make her carry them into the cells that she will
soon be building.
But in the meantime the future parasites must hold
tight to the fleece of their hostess, in spite of her rapid
flights among the flowers, in spite of her rubbing against
the walls of the galleries when she enters to take shelter,
and in spite, above all, of the brushing which she must
often give herself with her feet, to dust herself and keep
spick and span. We were wondering a little time ago
what the dangerous, shifting thing could be on which
the grub would have to establish itself. That thing is
the hair of a Bee who makes a thousand rapid journeys,
now diving into her narrow galleries, now forcing her
way down the tight throat of a flower.
We can now quite understand the use of the two
spikes, which close together and are able to take hold
of hair more easily than the most delicate tweezers. We
[167]
FABRE'S BOOK OF INSECTS
can see the full value of the sticky liquid that helps the
tiny creature to hold fast; and we can realise that the
elastic probes and bristles on the legs serve to penetrate
the Bee's down and anchor the grub in position. The
more one considers this arrangement, which seems so
useless as the grub drags itself laboriously over a smooth
surface, the more does one marvel at all the machinery
which this fragile creature carries about to save it from
falling during its adventurous rides.
Ill
THE SECOND ADVENTURE
One 2 1st of May I went to Carpentras, determined to
see, if possible, the entrance of the Sitaris into the Bee's
cells.
The works were in full swing. In front of a high ex-
panse of earth a swarm of Bees, stimulated by the sun,
was dancing a crazy ballet. From the tumultuous heart
of the cloud rose a monotonous, threatening murmur,
while my bewildered eye tried to follow the movements
of the throng. Quick as a lightning-flash thousands of
Anthophorse were flying hither and thither in search of
booty: thousands of others, also, were arriving, laden
with honey, or with mortar for their building.
At that time I knew comparatively little about these
insects. It seemed to me that any one who ventured
[168]
THE ADVENTURES OF A GRUB
into the swarm, or — above all — who laid a rash hand
on the Bees' dwellings, would instantly be stabbed by
a thousand stings. I had once observed the combs of the
Hornet too closely; and a shiver of fear passed through
me.
Yet, to find out what I wished to know, I must needs
penetrate that fearsome swarm; I must stand for whole
hours, perhaps all day, watching the works I intended
to upset; lens in hand, I must examine, unmoved amid
the whirl, the things that were happening in the cells.
Moreover, the use of a mask, of gloves, of a covering
of any kind, was out of the question, for my fingers and
eyes must be absolutely free. No matter: even though
I should leave the Bee's nest with my face swollen
beyond recognition, I was determined that day to solve
the problem that had puzzled me too long.
Having caught a few stray Anthophorae with my net,
I satisfied myself that the Sitaris-larvse were perched, as
I expected, on the Bees.
I buttoned my coat tightly and entered the heart of
the swarm. With a few blows of the mattock I secured
a lump of earth, and to my great surprise found myself
uninjured. A second expedition, longer than the first,
had the same result: not a Bee touched me with her
sting. After this I remained permanently in front of
the nest, removing lumps of earth, spilling the honey,
[169]
FABRE'S BOOK OF INSECTS
and crushing the Bees, without arousing anything worse
than a louder hum. For the Anthophora is a pacific
creature. When disturbed in the cells it leaves them
hastily and escapes, sometimes even mortally wounded,
without using its venomous sting except when it is
seized and handled.
Thanks to this unexpected lack of spirit in the Mason-
bee, I was able for hours to investigate her cells at my
leisure, seated on a stone in the midst of the murmuring
and distracted swarm, without receiving a single sting,
though I took no precautions whatever. Country folk,
happening to pass and seeing me seated thus calmly
amid the Bees, stopped aghast to ask me if I had be-
witched them.
In this way I examined the cells. Some were still
open, and contained only a more or less complete store
of honey. Others were closely sealed with an earthen
lid. The contents of these varied greatly. Sometimes
I found the larva of a Bee; sometimes another, fatter
kind of larva; at other times honey with an egg floating
on the surface. The egg was of a beautiful white, and
was shaped like a cylinder with a slight curve, a fifth
or sixth of an inch in length — the egg of the Anthophora.
In a few cells I found this tgg floating all alone on the
surface of the honey : in others, very many others, I saw,
[170]
THE ADVENTURES OF A GRUB
lying on the Bee's egg as though on a sort of raft, a young
Sitaris-grub. Its shape and size were those of the crea-
ture when it is hatched. Here, then, was the enemy
within the gates.
When and how did it get in? In none of the cells
was I able to detect any chink by which it could have
entered: they were all sealed quite tightly. The para-
site must have established itself in the honey-warehouse
before the warehouse was closed. On the other hand,
the open cells, full of honey but as yet without an egg,
never contain a Sitaris. The grub must therefore gain
admittance either while the Bee is laying the egg, or
else afterwards, while she is busy plastering up the door.
My experiments have convinced me that the Sitaris
enters the cell in the very second when the egg is laid
on the surface of the honey.
If I take a cell full of honey, with an egg floating in
it, and place it in a glass tube with some Sitaris-grubs,
they very rarely venture inside it. They cannot reach
the raft in safety: the honey that surrounds it is too
dangerous. If one of them by chance approaches the
honey it tries to escape as soon as it sees the sticky nature
of the stuff under its feet. It often ends by falling back
into the cell, where it dies of suffocation. It is therefore
certain that the grub does not leave the fleece of the Bee
[171]
FABRE'S BOOK OF INSECTS
when the latter is in her cell or near it, in order to make
a rush for the honey; for this honey would inevitably
cause its death, if it so much as touched the surface.
We must remember that the young Sitaris which is
found in a closed cell is always placed on the egg of the
Bee. This egg not only serves as a raft for the tiny
creature floating on a very treacherous lake, but also
provides it with its first meal. To get at this egg, in
the centre of the lake of honey, to reach this raft which
is also its first food, the young grub must somehow con-
trive to avoid the fatal touch of the honey.
There is only one way in which this can be done. The
clever grub, at the very moment when the Bee is laying
her egg, slips off the Bee and on to the egg, and with
it reaches the surface of the honey. The egg is too
small to hold more than one grub, and that is why we
never find more than one Sitaris in a cell. Such a per-
formance on the part of a grub seems extraordinarily
inspired — but then the study of insects constantly gives
us examples of such inspiration.
When dropping her egg upon the honey, then, the
Anthophora at the same time drops into her cell the
mortal enemy of her race. She carefully plasters the
lid which closes the entrance to the cell, and all is done.
A second cell is built beside it, probably to suffer the
same fate ; and so on until all the parasites sheltered by
[172]
THE ADVENTURES OF A GRUB
her fleece are comfortably housed. Let us leave the
unhappy mother to continue her fruitless task, and turn
our attention to the young larva which has so cleverly
secured for itself board and lodging.
Let us suppose that we remove the lid from a cell in
which the egg, recently laid, supports a Sitaris-grub.
The egg is intact and in perfect condition. But now
the work of destruction begins. The grub, a tiny black
speck which we see running over the white surface of
the egg, at last stops and balances itself firmly on its
six legs; then, seizing the delicate skin of the egg with
the sharp hooks of its mandibles, it tugs at it violently
till it breaks and spills the contents. These contents
the grub eagerly drinks up. Thus the first stroke of the
parasite's mandibles is aimed at the destruction of the
Bee's egg.
This is a very wise precaution on the part of the
Sitaris-grub I It will have to feed on the honey in the
cell: the Bee's grub which would come out of the egg
would also require the honey: there is not enough for
two. So — quick! — a bite at the egg, and the difficulty
is removed.
Moreover, another reason for the destruction of the
egg is that special tastes compel the young Sitaris to
make its first meals of it. The tiny creature begins by
greedily drinking the juices which the torn wrapper of
[173]
FABRE'S BOOK OF INSECTS
the egg allows to escape. For several days it continues
to rip the envelope gradually open, and to feed on the
liquid that trickles from it. Meanwhile it never touches
the honey that surrounds it. The Bee's egg is abso-
lutely necessary to the Sitaris-grub, not merely as a
boat, but also as nourishment.
At the end of a week the egg is nothing but a dry
skin. The first meal is finished. The Sitaris-grub,
which is now twice as large as before, splits open along
the back, and through this slit the second form of this
singular Beetle falls on the surface of the honey. Its
cast skin remains on the raft, and will presently dis-
appear with it beneath the waves of honey.
Here ends the history of the first form adopted by the
Sitaris.
[174]
CHAPTER XII
THE CRICKET
THE HOUSEHOLDER
THE Field Cricket, the inhabitant of the
meadows, is almost as famous as the Cicada,
and figures among the limited but glorious
number of the classic insects. He owes this honour to
his song and his house. One thing alone is lacking to
complete his renown. The master of the art of making
animals talk, La Fontaine, gives him hardly two lines.
Florian, the other French writer of fables, gives us a
story of a Cricket, but it lacks the simplicity of truth
and the saving salt of humour. Besides, it represents
the Cricket as discontented, bewailing his condition!
This is a preposterous idea, for all who have studied
him know, on the contrary, that he is very well pleased
with his own talent and his own burrow. And indeed,
at the end of the story, Florian makes him admit:
"My snug little home is a place of delight;
If you want to live happy, live hidden from sight."
[•75]
FABRE'S BOOK OF INSECTS
I find more force and truth in some verses by a friend
of mine, of which these are a translation :
Among the beasts a tale is told
How a poor Cricket ventured nigh
His door to catch the sun's warm gold
And saw a radiant Butterfly.
She passed with tails thrown proudly back
And long gay rows of crescents blue,
Brave yellow stars and bands of black.
The lordliest Fly that ever flew.
"Ah, fly away," the hermit said,
"Daylong among your flowers to roam;
Nor daisies white nor roses red
Will compensate my lowly home."
True, all too true ! There came a storm
And caught the Fly within its flood,
Staining her broken velvet form
And covering her wings with mud.
The Cricket, sheltered from the rain.
Chirped, and looked on with tranquil eye;
For him the thunder pealed in vain,
The gale and torrent passed him by.
Then shun the world, nor take your fill
Of any of its joys or flowers;
A lowly fire-side, calm and still.
At least will grant you tearless hours I ^
1 English transalation by Mr Stephen M'Kenna.
[176]
THE CRICKET
There I recognise my Cricket. I see him curling his
antennae on the threshold of his burrow, keeping him-
self cool in front and warm at the back. He is not
jealous of the Butterfly; on the contrary, he pities her,
with that air of mocking commiseration we often see in
those who have houses of their own when they are talk-
ing to those who have none. Far from complaining, he
is very well satisfied both with his house and his violin.
He is a true philosopher : he knows the vanity of things
and feels the charm of a modest retreat away from the
riot of pleasure-seekers.
Yes, the description is about right, as far as it goes.
But the Cricket is still waiting for the few lines needed
to bring his merits before the public; and since La Fon-
taine neglected him, he will have to go on waiting a
long time.
To me, as a naturalist, the important point in the two
fables is the burrow on which the moral is founded.
Florian speaks of the snug retreat; the other praises his
lowly home. It is the dwelling, therefore, that above
all compels attention, even that of the poet, who as a
rule cares little for realities.
In this matter, indeed, the Cricket is extraordinary.
Of all our insects he is the only one who, when full-
grown, possesses a fixed home, the reward of his own
industry. During the bad season of the year, most of
[177]
FABRE'S BOOK OF INSECTS
the others burrow or skulk in some temporary refuge,
a refuge obtained free of cost and abandoned without
regret. Several of them create marvels with a view to
settling their family: cotton satchels, baskets made of
leaves, towers of cement. Some live permanently in
ambush, lying in wait for their prey. The Tiger-beetle,
for instance, digs himself a perpendicular hole, which
he stops up with his flat, bronze head. If any other
insect steps on this deceptive trap-door it immediately
tips up, and the unhappy wayfarer disappears into the
gulf. The Ant-lion makes a slanting funnel in the
sand. Its victim, the Ant, slides down the slant and
is then stoned, from the bottom of the funnel, by the
hunter, who turns his neck into a catapult. But these
are all temporary refuges or traps.
The laboriously constructed home, in which the insect
settles down with no intention of moving, either in the
happy spring or in the woeful winter season; the real
manor-house, built for peace and comfort, and not as
a hunting-box or a nursery- — this is known to the Cricket
alone. On some sunny, grassy slope he is the owner of
a hermitage. While all the others lead vagabond lives,
sleeping in the open air or under the casual shelter of
a dead leaf or a stone, or the pealing bark of an old tree,
he is a privileged person with a permanent address.
The making of a home is a serious problem. It has
[178]
THE CRICKET
been solved by the Cricket, by the Rabbit, and lastly by
man. In my neighbourhood the Fox and the Badger
have holes, which are largely formed by the irregularities
of the rock. A few repairs, and the dug-out is com-
pleted. The Rabbit is cleverer than these, for he builds
his house by burrowing wherever he pleases, when there
is no natural passage that allows him to settle down free
of all trouble.
The Cricket is cleverer than any of them. He scorns
chance refuges, and always chooses the site of his home
carefully, in well-drained ground, with a pleasant sunny
aspect. He refuses to make use of ready-made caves
that are inconvenient and rough: he digs every bit of
his villa, from the entrance-hall to the back-room.
I see no one above him, in the art of house-building,
except man; and even man, before mixing mortar to
hold stones together, or kneading clay to coat his hut
of branches, fought with wild beasts for a refuge in the
rocks. Why is it that a special instinct is bestowed on
one particular creature'? Here is one of the humblest
of creatures able to lodge himself to perfection. He
has a home, an advantage unknown to many civilised
beings; he has a peaceful retreat, the first condition of
comfort; and no one around him is capable of settling
down. He has no rivals but ourselves.
Whence does he derive this gift'? Is he favoured
[179]
FABRE'S BOOK OF INSECTS
with special tools? No, the Cricket is not an expert
in the art of digging; in fact, one is rather surprised at
the result when one considers the feebleness of his means.
Is a home a necessity to him, on account of an excep-
tionally delicate skin*? No, his near kinsmen have skins
as sensitive as his, yet do not dread the open air at all.
Is the house-building talent the result of his anatomy *?
Has he any special organ that suggests it? No: in my
neighbourhood there are three other Crickets who are so
much like the Field Cricket in appearance, colour, and
structure, that at the first glance one would take them
for him. Of these faithful copies, not one knows how
to dig himself a burrow. The Double-spotted Cricket
inhabits the heaps of grass that are left to rot in damp
places; the Solitary Cricket roams about the dry clods
turned up by the gardener's spade ; the Bordeaux Cricket
is not afraid to make his way into our houses, where he
sings discreetly, during August and September, in some
cool, dark spot.
There is no object in continuing these questions: the
answer would always be No. Instinct never tells us its
causes. It depends so little on an insect's stock of tools
that no detail of anatomy, nothing in the creature's
formation, can explain it to us or make us foresee it.
These four similar Crickets, of whom only one can
[180]
THE CRICKET
burrow, are enough to show us our ignorance of the
origin of instinct.
Who does not know the Cricket's house *? Who has
not, as a child playing in the fields, stopped in front of
the hermit's cabin? However light your footfall, he
has heard you coming, and has abruptly withdrawn to
the very bottom of his hiding-place. When you arrive,
the threshold of the house is deserted.
Every one knows the way to bring out the skulker.
You insert a straw and move it gently about the burrow.
Surprised at what is happening above, the tickled and
teased Cricket ascends from his back room; he stops in the
passage, hesitates, and waves his delicate antennae inquir-
ingly. He comes to the light, and, once outside, he is
easy to catch, since these events have puzzled his poor
head. Should he be missed at the first attempt he may be-
come suspicious and refuse to appear. In that case he can
be flooded out with a glass of water.
Those were adorable times when we were children, and
hunted Crickets along the grassy paths, and put them in
cages, and fed them on a leaf of lettuce. They all come
back to me to-day, those times, as I search the burrows for
subjects to study. They seem like yesterday when my
companion, little Paul, an expert in the use of the straw,
springs up suddenly after a long trial of skill and
[181]
FABRE'S BOOK OF INSECTS
patience, and cries excitedly: "I've got him I I've got
him!"
Quick, here's a bag I In you go, my little Cricket!
You shall be petted and pampered, but you must teach us
something, and first of all you must show us your house.
II
HIS HOUSE
It is a slanting gallery in the grass, on some sunny
bank which soon dries after a shower. It is nine inches
long at most, hardly as thick as one's finger, and straight
or bent according to the nature of the ground. As a
rule, a tuft of grass half conceals the home, serving as
a porch and throwing the entrance discreetly into shadow.
When the Cricket goes out to browse upon the surround-
ing turf he does not touch this tuft. The gently
sloping threshold, carefully raked and swept, extends
for some distance ; and this is the terrace on which, when
everything is peaceful round about, the Cricket sits and
scrapes his fiddle.
The inside of the house is devoid of luxury, with
bare and yet not coarse walls. The inhabitant has
plenty of leisure to do away with any unpleasant rough-
ness. At the end of the passage is the bedroom, a little
more carefully smoothed than the rest, and slightly
wider. All said, it is a very simple abode, exceedingly
[182]
THE CRICKET
clean, free from damp, and conforming to the rules of
hygiene. On the other hand, it is an enormous under-
taking, a gigantic tunnel, when we consider the modest
tools with which the Cricket has to dig. If we wish to
know how he does it, and when he sets to work, we must
go back to the time when the egg is laid.
The Cricket lays her eggs singly in the soil, like the
Decticus, at a depth of three-quarters of an inch. She
arranges them in groups, and lays altogether about five
or six hundred. The egg is a little marvel of mechan-
ism. After the hatching it appears as an opaque white
cylinder, with a round and very regular hole at the top.
To the edge of this hole is fastened a cap, like a lid. In-
stead of bursting open anyhow under the thrusts of the
larva within, it opens of its own accord along a circular
line — a specially prepared line of least resistance.
About a fortnight after the egg is laid, two large,
round, rusty-black dots darken the front end. A little
way above these two dots, right at the top of the cylinder,
you see the outline of a thin circular swelling. This
is the line where the shell is preparing to break open.
Soon the transparency of the egg allows one to see the
delicate markings of the tiny creature's segments. Now
is the time to be on the watch, especially in the morning.
Fortune loves the persevering, and if we pay constant
visits to the eggs we shall be rewarded. All round the
swelling, where the resistance of the shell has gradually
[183]
FABRE'S BOOK OF INSECTS
been overcome, the end of the egg becomes detached.
Being pushed back by the forehead of the little creature
within, it rises and falls to one side like the top of a tiny
scent-bottle. The Cricket pops out like a Jack-in-the-
box.
When he is gone the shell remains distended, smooth,
intact, pure white, with the cap or lid hanging from the
opening. A bird's egg breaks clumsily under the blows
of a wart that grows for the purpose at the end of the
Chick's beak; the Cricket's egg is more ingeniously made,
and opens like an ivory case. The thrust of the crea-
ture's head is enough to work the hinge.
I said above that, when the lid is lifted, a young
Cricket pops out; but this is not quite accurate. What
appears is the swaddled grub, as yet unrecognisable in
a tight-fitting sheath. The Decticus, you will remember,
who is hatched in the same way under the soil, wears a
protective covering during his journey to the surface.
The Cricket is related to the Decticus, and therefore
wears the same livery, although in point of fact he does
not need it. The egg of the Decticus remains under-
ground for eight months, so the poor grub has to fight
its way through soil that has grown hard, and it therefore
needs a covering for its long shanks. But the Cricket
is shorter and stouter, and since its egg is only in the
ground for a few days it has nothing worse than a
powdery layer of earth to pass through. For these
[184]
THE CRICKET
reasons it requires no overall, and leaves it behind in
the shell.
As soon as he is rid of his swaddling-clothes the young
Cricket, pale all over, almost white, begins to battle with
the soil overhead. He hits out with his mandibles; he
sweeps aside and kicks behind him the powdery earth,
which offers no resistance. Very soon he is on the sur-
face, amidst the joys of the sunlight and the perils of
conflict with his fellow-creatures — poor feeble mite that
he is, hardly larger than a Flea.
By the end of twenty-four hours he has turned into
a magnificent blackamoor, whose ebon hue vies with that
of the full-grown insect. All that remains of his origi-
nal pallor is a white sash that girds his chest. Very
nimble and alert, he sounds the surrounding air with his
long, quivering antennae, and runs and jumps about with
great impetuosity. Some day he will be too fat to in-
dulge is such antics.
And now we see why the mother Cricket lays so many
eggs. It is because most of the young ones are doomed
to death. They are massacred in huge numbers by other
insects, and especially by the little Grey Lizard and the
Ant. The latter, loathsome freebooter that she is,
hardly leaves me a Cricket in my garden. She snaps
up the poor little creatures and gobbles them down at
frantic speed.
Oh, the execrable wretch! And to think that we
[185]
FABRE'S BOOK OF INSECTS
place the Ant in the front rank of insects I Books are
written in her honour, and the stream of praise never
runs dry. The naturalists hold her in great esteem; and
add daily to her fame. It would seem that with animals,
as with men, the surest way to attract attention is to
do harm to others.
Nobody asks about the Beetles who do such valuable
work as scavengers, whereas everybody knows the Gnat,
that drinker of men's blood; the Wasp, that hot-tempered
swashbuckler, with her poisoned dagger; and the Ant,
that notorious evil-doer who, in our southern villages,
saps and imperils the rafters of a dwelling as cheerfully
as she eats a fig.
The Ant massacres the Crickets in my garden so
thoroughly that I am driven to look for them outside the
enclosure. In August, among the fallen leaves, where
the grass has not been wholly scorched by the sun, I
find the young Cricket, already rather big, and now black
all over, with not a vestige of his white girdle remaining.
At this period of his life he is a vagabond : the shelter of
a dead leaf or a flat stone is enough for him.
Many of those who survived the raids of the Ants now
fall victims to the Wasp, who hunts down the wanderers
and stores them underground. If they would but dig
their dwellings a few weeks before the usual time they
would be saved; but they never think of it. They are
faithful to their ancient customs.
[186]
THE CRICKET
It is at the close of October, when the first cold weather
threatens, that the burrow is taken in hand. The work
is very simple, if I may judge by my observation of the
caged insect. The digging is never done at a bare point
in the pan, but always under the shelter of some withered
lettuce-leaf, a remnant of the food provided. This
takes the place of the grass tuft that seems indispensable
to the secrecy of the home.
The miner scrapes with his fore-legs, and uses the
pincers of his mandibles to pull out the larger bits of
gravel. I see him stamping with his powerful hind-
legs, furnished with a double row of spikes; I see him
raking the rubbish, sweeping it backwards and spreading
it slantwise. There you have the whole process.
The work proceeds pretty quickly at first. In the
yielding soil of my cages the digger disappears under-
ground after a spell that lasts a couple of hours. He
returns to the entrance at intervals, always backwards
and always sweeping. Should he be overcome with
fatigue he takes a rest on the threshold of his half-
finished home, with his head outside and his antenna
waving feebly. He goes in again, and resumes work
with pinchers and rakes. Soon the periods of rest be-
come longer, and wear out my patience.
The most urgent part of the work is done. Once
the hole is a couple of inches deep, it suffices for the needs
of the moment. The rest will be a long affair, carried
[187]
FABRE'S BOOK OF INSECTS
out in a leisurely way, a little one day and a little the
next: the hole will be made deeper and wider as the
weather grows colder and the insect larger. Even in
winter, if the temperature be mild and the sun shining
on the entrance to the dwelling, it is not unusual to see
the Cricket shooting out rubbish. Amid the joys of
spring the upkeep of the building still continues. It is
constantly undergoing improvements and repairs until
the owner's death.
When April ends the Cricket's song begins; at first
in rare and shy solos, but soon in a general symphony in
which each clod of turf boasts its performer. I am more
than inclined to place the Cricket at the head of the
spring choristers. In our waste-lands, when the thyme
and lavender are gaily flowering, the Crested Lark rises
like a lyrical rocket, his throat swelling with notes, and
from the sky sheds his sweet music upon the fallows.
Down below the Crickets chant the responses. Their
song is monotonous and artless, but well suited in its
very lack of art to the simple gladness of reviving life.
It is the hosanna of the awakening, the sacred alleluia
understood by swelling seed and sprouting blade. In
this duet I should award the palm to the Cricket. His
numbers and his unceasing note deserve it. Were the
Lark to fall silent, the fields blue-grey with lavender,
swinging its fragrant censors before the sun, would still
[188]
THE CRICKET
receive from this humble chorister a solemn hymn of
praise.
Ill
HIS MUSICAL-BOX
In steps Science, and says to the Cricket bluntly:
"Show us your musical-box."
Like all things of real value, it is very simple. It is
based on the same principle as that of the Grasshoppers :
a bow with a hook to it, and a vibrating membrane.
The right wing-case overlaps the left and covers it almost
completely, except where it folds back sharply and en-
cases the insect's side. It is the opposite arrangement to
that which we find in the Green Grasshopper, the Decti-
cus, and their kinsmen. The Cricket is right-handed,
the others left-handed.
The two wing-cases are made in exactly the same way.
To know one is to know the other. They lie flat on the
insect's back, and slant suddenly at the side in a right-
angled fold, encircling the body with a delicately veined
pinion.
If you hold one of these wing-cases up to the light
you will see that is it a very pale red, save for two large
adjoining spaces; a larger, triangular one in front, and
a smaller, oval one at the back. They are crossed by
[189]
FABRE'S BOOK OF INSECTS
faint wrinkles. These two spaces are the sounding-
boards, or drums. The skin is finer here than elsewhere,
and transparent, though of a somewhat smoky tint.
At the hinder edge of the front part are two curved,
parallel veins, with a cavity between them. This cavity
contains five or six little black wrinkles that look like
the rungs of a tiny ladder. They supply friction : they
intensify the vibration by increasing the number of
points touched by the bow.
On the lower surface one of the two veins that sur-
round the cavity of the rungs becomes a rib cut into the
shape of a hook. This is the bow. It is provided with
about a hundred and fifty triangular teeth of exquisite
geometrical regularity.
It is a fine instrument indeed. The hundred and
fifty teeth of the bow, biting into the rungs of the oppo-
site wing-case, set the four drums in motion at one and
the same time, the lower pair by direct friction, the upper
pair by the shaking of the friction-apparatus. What a
rush of sound I The Cricket with his four drums throws
his music to a distance of some hundreds of yards.
He vies with the Cicada* in shrillness, without having
the latter's disagreeable harshness. Andt better still:
this favoured creature knows how to modulate his song.
The wing-cases, as I said, extend over each side in a wide
fold. These are the dampers which, lowered to a
greater or less depth, alter the intensity of the sound.
[190]
THE CRICKET
According to the extent of their contact with the soft
body of the Cricket they allow him to sing gently at one
time and fortissimo at another.
The exact similarity of the two wing-cases is worthy
of attention. I can see clearly the function of the upper
bow, and the four sounding-spaces which sets it in
motion ; but what is the good of the lower one, the bow
on the left wing? Not resting on anything, it has
nothing to strike with its hook, which is as carefully
toothed as the other. It is absolutely useless, unless the
apparatus can invert the order of its two parts, and place
that above which is below. If that could be done, the
perfect symmetry of the instrument is such that the
mechanism would be the same as before, and the insect
would be able to play with the bow that is at present
useless. The lower fiddlestick would become the upper,
and the tune would be the same.
I suspected at first that the Cricket could use both
bows, or at least that there were some who were per-
manently left-handed. But observation has convinced
me of the contrary. All the Crickets I have examined—
and they are many— without a single exception carried
the right wing-case above the left.
I even tried to bring about by artificial means what
Nature refused to show me. Using my forceps, very
gently of course, and without straining the wing-cases,
I made these overlap the opposite way. It is easily done
[191]
FABRE'S BOOK OF INSECTS
with a little skill and patience. Everything went well :
there was no dislocation of the shoulders, the membranes
were not creased.
I almost expected the Cricket to sing, but I was soon
undeceived. He submitted for a few moments; but
then, finding himself uncomfortable, he made -an effort
and restored his instrument to its usual position. In
vain I repeated the operation: the Cricket's obstinacy
triumphed over mine.
Then I thought I would make the attempt while the
wing-cases were quite new and plastic, at the moment
when the larva casts its skin. I secured one at the point
of being transformed. At this stage the future wings
and wing-cases form four tiny flaps, which, by their shape
and scantiness, and by the way they stick out in different
directions, remind me of the short jackets worn by the
Auvergne cheesemakers. The larva cast off these gar-
ments before my eyes.
The wing-cases developed bit by bit, and opened out.
There was no sign to tell me which would overlap the
other. Then the edges touched : a few moments longer
and the right would be over the left. This was the time
to intervene.
With a straw I gently changed the position, bringing
the left edge over the right. In spite of some protest
from the insect I was quite successful: the left wing-
case pushed forward, though only very little. Then I
[192]
THE CRICKET
left it alone, and gradually the wing-cases matured in
the inverted position. The Cricket was left-handed.
I expected soon to see him wield the fiddlestick which
the members of his family never employ.
On the third day he made a start. A few brief grating
sounds were heard — the noise of a machine out of gear
shifting its parts back into their proper order. Then the
tune began, with its accustomed tone and rhythm.
Alas, I had been over-confident in my mischievous
straw! I thought I had created a new type of instru-
mentalist, and I had obtained nothing at all! The
Cricket was scraping with his right fiddlestick, and
always would. With a painful effort he had dislocated
his shoulders, which I had forced to harden in the wrong
way. He had put back on top that which ought to be
on top, and underneath that which ought to be under-
neath. My sorry science tried to make a left-handed
player of him. He laughed at my devices, and settled
down to be right-handed for the rest of his life.
Enough of the instrument; let us listen to the music.
The Cricket sings on the threshold of his house, in the
cheerful sunshine, never indoors. The wing-cases utter
their cri-cri in a soft tremolo. It is full, sonorous, nicely
cadenced, and lasts indefinitely. Thus are the leisures
of solitude beguiled all through the spring. The hermit
at first sings for his own pleasure. Glad to be alive, he
chants the praises of the sun that shines upon him, the
[193]
FABRE'S BOOK OF INSECTS
grass that feeds him, the peaceful retreat that harbours
him. The first object of his bow is to hymn the pleasures
of life.
Later on he plays to his mate. But, to tell the truth,
his attention is rewarded with little gratitude; for in the
end she quarrels with him ferociously, and unless he
takes to flight she cripples him — and even eats him more
or less. But indeed, in any case he soon dies. Even if
he escapes his pugnacious mate, he perishes in June.
We are told that the music-loving Greeks used to keep
Cicadae in cages, the better to enjoy their singing. I
venture to disbelieve the story. In the first place the
harsh clicking of the Cicadae, when long continued at
close quarters, is a torture to ears that are at all delicate.
The Greeks' sense of hearing was too well trained to take
pleasure in such raucous sounds away from the general
concert of the fields, which is heard at a distance.
In the second place it is absolutely impossible to bring
up Cicadae in captivity, unless we cover over a whole
olive-tree or plane-tree. A single day spent in a
cramped enclosure would make the high-flying insect die
of boredom.
Is it not possible that people have confused the Cricket
with the Cicada, as they also do the Green Grasshopper*?
With the Cricket they would be quite right. He is one
who bears captivity gaily: his stay-at-home ways pre-
[194]
THE CRICKET
dispose him to it. He lives happily and whirrs without
ceasing in a cage no larger than a man's fist, provided
that he has his lettuce-leaf every day. Was it not he
whom the small boys of Athens reared in little wire cages
hanging on a window-frame?
The small boys of Provence, and all the South, have
the same tastes. In the towns a Cricket becomes the
child's treasured possession. The insect, petted and
pampered, sings to him of the simple joys of the country.
Its death throws the whole household into a sort of
mourning.
The three other Crickets of my neighbourhood all
carry the same musical instrument as the Field Cricket,
with slight variation of detail. Their song is much
alike in all cases, allowing for differences of size. The
smallest of the family, the Bordeaux Cricket, sometimes
ventures into the dark corners of my kitchen, but his
song is so faint that it takes a very attentive ear to hear
it.
The Field Cricket sings during the sunniest hours of
the spring: during the still summer nights we have the
Italian Cricket. He is a slender, feeble insect, quite
pale, almost white, as beseems his nocturnal habits.
You are afraid of crushing him, if you so much as take
him in your fingers. He lives high in air, on shrubs of
every kind, or on the taller grasses; and he rarely de-
[195]
FABRE'S BOOK OF INSECTS
scends to earth. His song, the sweet music of the still,
hot evenings from July to October; begins at sunset and
continues for the best part of the night.
This song is known to everybody here in Provence, for
the smallest clump of bushes has its orchestra. The
soft, slow gri-i-i gri-i-i is made more expressive by a
slight tremolo. If nothing happens to disturb the insect
the sound remains unaltered; but at the least noise the
musician becomes a ventriloquist. You hear him quite
close, in front of you; and then, all of a sudden, you hear
him fifteen yards away. You move towards the sound.
It is not there: it comes from the original place. No,
it doesn't after all. Is it over there on the left, or does
it come from behind*? One is absolutely at a loss, quite
unable to find the spot where the music is chirping.
This illusion of varying distance is produced in two
ways. The sounds become loud or soft, open or muffled,
according to the exact part of the lower wing-case that
is pressed by the bow. And they are also modified by
the position of the wing-cases. For the loud sounds
these are raised to their full height: for the muffled
sounds they are lowered more or less. The pale Cricket
misleads those who hunt for him by pressing the edges of
his vibrating flaps against his soft body.
I know no prettier or more limpid insect-song than
his, heard in the deep stillness of an August evening.
How often have I lain down on the ground among the
[196]
THE CRICKET
rosemary bushes of my harmas, to listen to the delightful
concert I
The Italian Cricket swarms in my enclosure. Every
tuft of red-flowering rock-rose has its chorister; so has
every clump of lavender. The bushy arbutus-shrubs,
the turpentine-trees, all become orchestras. And in its
clear voice, so full of charm, the whole of this little
world, from every shrub and every branch, sings of the
gladness of life.
High up above my head the Swan stretches its great
cross along the Milky Way: below, all round me, the
insect's symphony rises and falls. Infinitesimal life
telling its joys makes me forget the pageant of the stars.
Those celestial eyes look down upon me, placid and cold,
but do not stir a fibre within me. Why? They lack
the great secret — life. Our reason tells us, it is true,
that those suns warm worlds like ours; but when all is
said, this belief is no more than a guess, it is not a
certainty.
In your company, on the contrary, O my Cricket, I
feel the throbbing of life, which is the soul of our lump
of clay; and that is why, under my rosemary-hedge, I
give but an absent glance at the constellation of the
Swan and devote all my attention to your serenade! A
living speck— the merest dab of life— capable of
pleasure and pain, is far more interesting to me than all
the immensities of mere matter.
[197]
CHAPTER XIII
THE SISYPHUS
YOU are not tired, I hope, of hearing about the
Scavenger Beetles with a talent for making
balls. I have told you of the Sacred Beetle
and of the Spanish Copris, and now I wish to say a few
words of yet another of these creatures. In the insect
world we meet with a great many model mothers : it is
only fair, for once to draw attention to a good father.
Now a good father is rarely seen except among the
higher animals. The bird is excellent in this respect,
and the furred folk perform their duties honourably.
Lower in the scale of living creatures the father is
generally indifferent to his family. Very few insects
are exceptions to this rule. This heartlessness, which
would be detestable in the higher ranks of the animal
kingdom, where the weakness of the young demands pro-
longed care, is excusable among insect fathers. For the
robustness of the new-born insect enables it to gather
its food unaided, provided it be in a suitable place.
When all that the Pieris need do for the safety of the
race is to lay her eggs on the leaves of a cabbage, of what
use would a father's care be*? The mother's botanical
[198]
THE SISYPHUS
instinct needs no assistance. At laying-time the other
parent would be in the way.
Most insects adopt this simple method of upbringing.
They merely choose a dining-room which will be the
home of the family once it is hatched, or else a place that
will allow the young ones to find suitable fare for them-
selves. There is no need for the father in such cases.
He generally dies without lending the least assistance
in the work of setting up his offspring in life.
Things do not always happen, however, in quite such
a primitive fashion. There are tribes that provide a
dowry for their families, that prepare board and lodging
for them in advance. The Bees and Wasps in particu-
lar are masters in the industry of making cellars, jars,
and satchels, in which the ration of honey is hoarded:
they are perfect in the art of creating burrows stocked
with the game that forms the food of their grubs.
Well, this enormous labour, which is one of building
and provisioning combined, this toil in which the insect's
whole life is spent, is done by the mother alone. It
wears her out ; it utterly exhausts her. The father drunk
with sunlight, stands idle at the edge of the workyard,
watching his plucky helpmate at her job.
Why does he not lend the mother a helping hand'? It
is now or never. Why does he not follow the example
of the Swallow couple, both of whom bring their bit of
straw, their blob of mortar to the building and their
[199]
FABRE'S BOOK OF INSECTS
Midge to the young ones ? He does nothing of the kind.
Possibly he puts forward his comparative weakness as an
excuse. It is a poor argument; for to cut a disk out of
a leaf, to scrape some cotton from a downy plant, to
collect a little bit of cement in muddy places would not
overtax his strength. He could very easily help, at any
rate as a labourer; he is quite fit to gather materials for
the mother, with her greater intelligence, to fit in place.
The real reason of his inactivity is sheer incapability.
It is strange that the most gifted of the industrial
insects should know nothing of a father's duties. One
would expect the highest talents to be developed in him
by the needs of the young; but he remains as dull-witted
as a Butterfly, whose family is reared at so small a cost.
We are baffled at every turn by the question: Why is
a particular instinct given to one insect and denied to
another?
It baffles us so thoroughly that we are extremely sur-
prised when we find in the scavenger the noble qualities
that are denied to the honey-gatherer. Various Scaven-
ger Beetles are accustomed to help in the burden of
housekeeping, and know the value of working in double
harness. The Geotrupes couple, for instance, prepare
their larva's food together : the father lends his mate the
assistance of his powerful press in the manufacture of
the tightly packed sausage-shaped ration. He is a splen-
[200]
THE SISYPHUS
did example of domestic habits, and one extremely sur-
prising amid the general egoism.
To this example my constant studies of the subject
have enabled me to add three others, all furnished by
the Guild of Scavengers.
One of them is the Sisyphus, the smallest and most
zealous of all our pill-rollers. He is the liveliest and
most agile of them all, and recks nothing of awkward
somersaults and headlong falls on the impossible roads
to which his obstinacy brings him back again and again.
It was in reference to these wild gymnastics that La-
treille gave him the name of Sisyphus.
As you know, that unhappy wretch of classical fame
had a terrible task. He was forced to roll a huge stone
uphill ; and each time he succeeded in toiling to the top
of the mountain the stone slipped from his grasp and
rolled to the bottom. I like this myth. It is the his-
tory of a good many of us. So far as I am concerned,
for half a century and more I have painfully climbed the
steep ascent, spending my strength recklessly in the
struggle to hoist up to safety that crushing burden, my
daily bread. Hardly is the loaf balanced when it slips
off, slides down, and is lost in the abyss.
The Sisyphus with whom we are now concerned knows
none of these bitter trials. Untroubled by the steep
slopes he gaily trundles his load, at one time bread for
himself, at another bread for his children. He is very
[201]
FABRE'S BOOK OF INSECTS
scarce in these parts; and I should never have managed
to secure a suitable number of subjects for my studies
had it not been for an assistant whom I have already
mentioned more than once.
I speak of my little son Paul, aged seven. He is my
enthusiastic companion on my hunting expeditions, and
knows better than any one of his age the secrets of the
Cicada, the Locust, the Cricket, and especially the
Scavenger Beetle. Twenty paces away his sharp eyes
will distinguish the real mound that marks a burrow
from casual heaps of earth. His delicate ears catch
the Grasshopper's faint song, which is quite unheard by
me. He lends me his sight and hearing; and I, in ex-
change, present him with ideas, which he receives
attentively.
Little Paul has his own insect-cages, in which the
Sacred Beetle makes pears for him; his own little garden,
no larger than a pocket-handkerchief, where he grows
beans, often digging them up to see if the tiny roots are
any longer; his forest plantation, in which stand four
oaks a hand's-breadth high, still furnished on one side
with the acorn that feeds them. It all makes a welcome
change from grammar, which gets on none the worse for
it.
When the month of May is near at hand Paul and I
get up early one morning — so early that we start without
our breakfast — and we explore, at the foot of the moun-
[202]
THE SISYPHUS
tain, the meadows where the flocks have been. Here
we find the Sisyphus. Paul is so zealous in his search
that we soon have a sufficient number of couples.
All that is needed for their well-being is a wire-gauze
cover, with a bed of sand and a supply of their food —
to obtain which we too turn scavengers. These creatures
are so small, hardly the size of a cherry-stone ! And so
curious in shape withal! A dumpy body, the hinder
end of which is pointed, and very long legs, resembling
a Spider's when outspread. The hind-legs are of amaz-
ing length, and are curved, which is most useful for
clasping and squeezing the pellet.
Soon the time comes for establishing the family.
With equal zeal father and mother alike take part in
kneading, carting, and stowing away the provisions for
the young ones. With the cleaver of the fore-legs a
morsel of the right size is cut from the food placed at
their disposal. The two insects work at the piece to-
gether, giving it little pats, pressing it, and shaping it
into a ball as large as a big pea.
As in the Sacred Beetle's workshop, the accurately
round shape is obtained without the mechanical trick
of rolling the ball. The material is modelled into a
sphere before it is moved, before it is even loosened from
its support. Here, once more, we have an expert in
geometry familiar with the best form for preserving
food.
[203]
FABRE'S BOOK OF INSECTS
The ball is soon ready. It must now, by vigorous
rolling, be given the crust which will protect the soft
stuff within from becoming too dry. The mother, who
can be recognised by her slightly larger size, harnesses
herself in the place of honour, in front. With her long
hind-legs on the ground and her fore-legs on the ball,
she hauls it towards her, backwards. The father pushes
behind in the reverse position, head downwards. It is
precisely the same method as that of the Sacred Beetle
when working in twos, but it has another object. The
Sisyphus team conveys a store of food for the grubs,
whereas the big pill-rollers trundle a banquet which they
themselves will eat up underground.
The couple start off along the ground. They have
no definite goal, but walk in a direct line, without regard
to the obstacles that lie in the way. In this backward
march the obstacles could not be avoided; but even if
they were seen the Sisyphus would not try to go round
them. For she even makes obstinate attempts to climb
the wire-work of my cage. This is an arduous and im-
possible task. Clawing the meshes of the gauze with
her hind-legs the mother pulls the load towards her;
then, putting her fore-legs round it, she holds it sus-
pended in air. The father, finding nothing to stand
upon, clings to the ball — encrusts himself in it, so to
speak, thus adding his weight to that of the lump, and
taking no further pains. The effort is too great to last.
[204]
THE SISYPHUS
The ball and its rider, forming one mass, fall to the floor.
The mother, from above, looks down for a moment in
surprise, and then drops to recover the load and renew
her impossible attempt to scale the side. After repeated
falls the climb is abandoned.
Even on level ground the carting is not carried on
without difficulty. At every moment the load swerves
on some mound made by a bit of gravel ; and the team
topple over and kick about, upside down. This is a
trifle, the merest trifle. These tumbles, which so often
fling the Sisyphus on his back, cause him no concern ; one
would even think he liked them. After all, the ball has
to be hardened and made of the right consistency. And
this being the case, bumps falls, and jolts are all part of
the programme. This mad steeple-chasing goes on for
hours.
At last the mother, regarding the work as completed,
goes off a little way in search of a suitable spot. The
father mounts guard, squatting on the treasure. If his
companion's absence be unduly long, he relieves his
boredom by spinning the ball nimbly between his up-
lifted hind legs. He treats his precious pellet as a
juggler treats his ball. He tests its perfect shape with
his curved legs, the branches of his compasses. No one
who sees him frisking in that jubilant attitude can doubt
his lively satisfaction — the satisfaction of a father as-
sured of his children's future.
[205]
FABRE'S BOOK OF INSECTS
"It is I," he seems to say, "I who kneaded this round
loaf, I who made this bread for my sons!"
And he lifts on high, for all to see, this magnificent
testimony to his industry.
Meanwhile the mother has chosen a site for the bur-
row. A shallow pit is made, a mere beginning of the
work. The ball is rolled near it. The father, that vigi-
lant guardian, does not let go, while the mother digs
with her legs and forehead. Soon the hollow is big
enough to hold the pellet. She insists on having it quite
close to her; she must feel it bobbing up and down be-
hind her, on her back, safe from parasites, before she
decides to go farther. She is afraid of what might
happen to it if it were left on the edge of the burrow
until the home were completed. There are plenty of
Midges and other such insects to grab it. One cannot
be too careful.
The ball therefore is inserted, half in and half out of
the partly-formed basin. The mother, underneath, gets
her legs round it and pulls: the father above, lets it
down gently, and sees that the hole is not choked up with
falling earth. All goes well. The digging is resumed
and the descent continues, always with the same caution;
one of the insects pulling the load, the other regulating
the drop and clearing away anything that might hinder
the operation. A few more efforts, and the ball dis-
appears underground with the two miners. What
[206]
THE SISYPHUS
follows for some time to come can only be a repetition of
what has already been done. We must wait half a day
or so.
If we keep careful watch we shall see the father come
up again to the surface by himself, and crouch in the
sand near the burrow. Detained below by duties in
which her companion can be of no assistance to her, the
mother usually postpones her appearance till the
morrow. At last she shows herself. The father leaves
the place where he was snoozing, and joins her. The re-
united couple go back to the spot where their food-stufFs
are to be found, and having refreshed themselves they
gather up more materials. The two then set to work
again. Once more they model, cart, and store the ball
together.
I am delighted with this constancy. That it is really
the rule I dare not declare. There must, no doubt, be
flighty, hckle Beetles. No matter : the little I have seen
gives me a high opinion of the domestic habits of the
Sisyphus.
It is time to inspect the burrow. At no great depth we
find a tiny niche, just large enough to allow the mother
to move round her work. The smallness of the chamber
tells us that the father cannot remain there for long.
When the studio is ready, he must go away to leave the
sculptress room to turn.
The contents of the cellar consist of a single ball, a
[207]
FABRE'S BOOK OF INSECTS
masterpiece of art. It is a copy of the Sacred Beetle's
pear on a very much reduced scale, its smallness making
the polish of the surface and the elegance of the curves
all the more striking. Its diameter, at the broadest
point, measures one-half to three-quarters of an inch.
One more observation about the Sisyphus. Six
couples under the wire-gauze cover gave me fifty-seven
pears containing one egg each — an average of over nine
grubs to each couple. The Sacred Beetle is far from
reaching this figure. To what cause are we to attribute
this large brood ? I can see but one : the fact that the
father works as well as the mother. Family burdens
that would exceed the strength of one are not too heavy
when there are two to bear them.
[208]
CHAPTER XIV
THE CAPRICORN
THE grub's home
AN eighteenth-century philosopher, Condillac,
describes an imaginary statue, organised like a
man, but with none of a man's senses. He
then pictures the effect of endowing it with the five
senses, one by one, and the first sense he gives it is that
of smell. The statue, having no sense but smell, in-
hales the scent of a rose, and out of that single impression
creates a whole world of ideas. In my youth I owed
some happy moments to that statue. I seemed to see it
come to life in that action of the nostrils, acquiring
memory, concentration, judgment, and other mental
qualities, even as still waters are aroused and rippled
by the impact of a grain of sand. I recovered from my
illusion under the teaching of my abler master the animal.
The Capricorn taught me that the problem is more ob-
scure than the Abbe Condillac led me to suppose.
When my winter supply of firewood is being prepared
for me with wedge and mallet, the woodman selects, by
my express orders, the oldest and most ravaged trunks
[209]
FABRE'S BOOK OF INSECTS
in his stack. My tastes bring a smile to his lips; he
wonders by what whimsy I prefer wood that is worm-
eaten to sound wood, which burns so much better. I
have my views on the subject, and the worthy man
submits to them.
A fine oak-trunk, seamed with scars and gashed with
wounds, contains many treasures for my studies. The
mallet drives home, the wedges bite, the wood splits;
and within, in the dry and hollow parts, are revealed
groups of various insects who are capable of living
through the cold season, and have here taken up their
winter quarters. In the low-roofed galleries built by
some Beetle the Osmia Bee has piled her cells one above
the other. In the deserted chambers and vestibules
Megachiles have arranged their leafy jars. In the live
wood, filled with juicy sap, the larva of the Capricorn,
the chief author of the oak's undoing, has set up its home.
Truly they are strange creatures, these grubs : bits of
intestines crawling about I In the middle of Autumn
I find them of two different ages. The older are almost
as thick as one's finger; the others hardly attain the
diameter of a pencil. I find, in addition, the pupa or
nymph more or less fully coloured, and the perfect insect
ready to leave the trunk when the hot weather comes
again. Life inside the wood, therefore, lasts for three
years.
How is this long period of solitude and captivity spent*?
[210]
THE CAPRICORN
In wandering lazily through the thickness of the oak, in
making roads whose rubbish serves as food. The horse
in the book of Job "swallows the ground" in a figure of
speech: the Capricorn's grub eats its way literally.
With its carpenter' s-gouge — a strong black mandible,
short and without notches, but scooped into a sharp-
edged spoon — it digs the opening of its tunnel. From
the piece cut out the grub extracts the scanty juices, while
the refuse accumulates behind him in heaps. The path
is devoured as it is made; it is blocked behind as it makes
way ahead.
Since this harsh work is done with the two gouges, the
two curved chisels of the mandibles, the Capricorn-grub
requires much strength in the front part of its body,
which therefore swells into a sort of pestle. The Bu-
prestis-grub, that other industrious carpenter, adopts a
similar form, and even exaggerates its pestle. The part
that toils and carves hard wood requires to be robust; the
rest of the body, which has but to follow after, continues
slim. The essential thing is that the implement of the
jaws should possess a solid support and powerful ma-
chinery. The Capricorn larva strengthens its chisels
with a stout, black, horny armour that surrounds the
mouth; yet, apart from its skull and its equipment of
tools, this grub has a skin as fine as satin and as white
as ivory. This dead white is caused by a thick layer of
grease, which one would not expect a diet of wood to
[211]
FABRE'S BOOK OF INSECTS
produce in the animal. True, it has nothing to do, at
every hour of the day and night, but gnaw. The quan-
tity of wood that passes into its stomach makes up for the
lack of nourishing qualities.
The grub's legs can hardly be called legs at all; they
are mere suggestions of the legs the full-grown insect
will have by and by. They are infinitesimal in size,
and of no use whatever for walking. They do not even
touch the supporting surface, being kept off it by the
plumpness of the chest. The organs by means of which
the animal progresses are something altogether different.
The grub of the Rose-chafer, with the aid of the hairs
and pad-like projections upon its spine, manages to
reverse the usual method of walking, and to wriggle
along on its back. The grub of the Capricorn is even
more ingenious : it moves at the same time on its back
and its stomach. To take the place of its useless legs it
has a walking apparatus almost like feet, which appear,
contrary to every rule, on the surface of its back.
On the middle part of its body, both above and below,
there is a row of seven four-sided pads, which the grub
can either expand or contract, making them stick out or
lie flat at will. It is by means of these pads that it
walks. When it wishes to move forwards it expands
the hinder pads, those on the back as well as those on the
stomach, and contracts its front pads. The swelling of
the hind pads in the narrow gallery fills up the space, and
[212]
THE CAPRICORN
gives the grub something to push against. At the same
time the flattening of the front pads, by decreasing the
size of the grub, allows it to slip forward and take half
a step. Then, to complete the step, the hind-quarters
must be brought up the same distance. With this object
the front pads fill out and provide support, while those
behind shrink and leave room for the grub to draw up
its hind-quarters.
With the double support of its back and stomach, with
alternate swellings and shrinkings, the animal easily
advances or retreats along its gallery, a sort of mould
which the contents fill without a gap. But if the pads
grip only on one side progress becomes impossible.
When placed on the smooth wood of my table the animal
wriggles slowly; it lengthens and shortens without pro-
gressing by a hair's breadth. Laid on the surface of a
piece of split oak, a rough, uneven surface due to the
gash made by the wedge, it twists and writhes, moves
the front part of its body very slowly from left to right
and right to left, lifts it a little, lowers it, and begins
again. This is all it can do. The rudimentary legs
remain inert and absolutely useless.
II
THE grub's sensations
Though the Capricorn-grub possesses these useless
legs, the .germs of future limbs, there is no sign of the
[213]
FABRE'S BOOK OF INSECTS
eyes with which the fully-developed insect will be richly
gifted. The larva has not the least trace of any organs
of sight. What would it do with sight, in the murky
thickness of a tree-trunk? Hearing is likewise absent.
In the untroubled silence of the oak's inmost heart the
sense of hearing would be superfluous. Where sounds
are lacking, of what use is the faculty of discerning
them?
To make the matter certain I carried out some ex-
periments. If split lengthwise the grub's abode becomes
a half-tunnel, in which I can watch the occupant's
doings. When left alone it alternately works for
a while, gnawing at its gallery, and rests for awhile,
fixed by its pads to the two sides of the tunnel. I took
advantage of these moments of rest to inquire into its
power of hearing. The banging of hard bodies, the ring
of metallic objects, the grating of a file upon a saw, were
tried in vain. The animal remained impassive: not a
wince, not a movement of the skin, no sign of awakened
attention. I succeeded no better when I scratched the
wood near it with a hard point, to imitate the sound of
some other grub at work in its neighbourhood. The in-
difference to my noisy tricks could be no greater in a
lifeless object. The animal is deaf.
Can it smell? Everything tells us that it cannot.
Scent is of assistance in the search for food. But the
Capricorn-grub need not go in quest of eatables. It
[214]
THE CAPRICORN
feeds on its home; it lives on the wood that gives it shel-
ter. Nevertheless I tested it. In a log of fresh cypress
wood I made a groove of the same width as that of the
natural galleries, and I placed the grub inside it. Cy-
press wood is strongly scented; it has the smell charac-
teristic of most of the pine family. This resinous scent,
so strange to a grub that lives always in oak, ought to
vex it, to trouble it; and it should show its displeasure
by some kind of commotion, some attempt to get away.
It did nothing of the kind: once it had found the right
position in the groove it went to the end, as far as it
could go, and made no further movement. Then I set
before it, in its usual channel, a piece of camphor.
Again no eifect. Camphor was followed by naphtha-
line. Still no result. I do not think I am going too far
when I deny the creature a sense of smell.
Taste is there no doubt. But such taste! The food
is without variety : oak, for three years at a stretch, and
nothing else. What can the grub's palate find to enjoy
in this monotonous fare? The agreeable sensation of
a fresh piece, oozing with sap; the uninteresting flavour
of an over-dry piece. These, probably, are the only
changes in the meal.
There remains the sense of touch, the universal pas-
sive sense common to all live flesh that quivers under the
goad of pain. The Capricorn-grub, therefore, is limited
to two senses, those of taste and touch, and both of these
[215]
FABRE'S BOOK OF INSECTS
it possesses only in a very small degree. It is very little
better off than Condillac's statue. The imaginary being
created by the philosopher had one sense only, that of
smell, equal in delicacy to our own; the real being, the
oak-eater has two, which are inferior even when put to-
gether to the one sense of the statue. The latter plainly
perceived the scent of a rose, and clearly distinguished
it from any other.
A vain wish has often come to me in my dreams : to
be able to think, for a few minutes, with the brain of
my Dog, or to see the world with the eyes of a Gnat.
How things would change in appearance! But they
would change much more if understood only with the
intellect of the grub. What has that incomplete crea-
ture learnt through its senses of touch and taste? Very
little; almost nothing. It knows that the best bits of
wood have a special kind of flavour, and that the sides
of a passage, when not carefully smoothed, are painful
to the skin. This is the limit of its wisdom. In com-
parison with this, the statue with the sensitive nostrils
was a marvel of knowledge. It remembered, compared,
judged, and reasoned. Can the Capricorn-grub remem-
ber? Can it reason? I described it a little time ago
as a bit of intestine that crawls about. This descrip-
tion gives an answer to these questions. The grub has
the sensations of a bit of intestine, no more and no less.
[216]
THE CAPRICORN
III
THE grub's foresight
And this half-alive object, this no thing -at-all, is cap-
able of marvellous foresight. It knows hardly anything
of the present, but it sees very clearly into the future.
For three years on end the larva wanders about in the
heart of the trunk. It goes up, goes down, turns to this
side and that; it leaves one vein for another of better
flavour, but without ever going too far from the inner
depths, where the temperature is milder than near the
surface, and greater safety reigns. But a day is at hand
when the hermit must leave its safe retreat and face the
perils of the outer world. Eating is not everything,
after all; we have to get out of this.
But how? For the grub, before leaving the trunk,
must turn into a long-horned Beetle. And though the
grub, being w^ell equipped with tools and muscular
strength, finds no difficulty in boring through the wood
and going where it pleases, it by no means follows that
the coming Capricorn has the same powers. The
Beetle's short spell of life must be spent in the open air.
Will it be able to clear itself a way of escape?
It is quite plain, at all events, that the Capricorn will
be absolutely unable to make use of the tunnel bored
by the grub. This tunnel is a very long and very irregu-
[217]
FABRE'S BOOK OF INSECTS
lar maze, blocked with great heaps of wormed wood.
It grows constantly smaller and smaller as it approaches
the starting-point, because the larva entered the trunk
as slim as a tiny bit of straw, whereas to-day it is as
thick as one's finger. In its three years' wanderings it
always dug its gallery to fit the size of its body. Evi-
dently the road of the larva cannot be the Capricorn's
way out. His overgrown antennae, his long legs, his
inflexible armour-plates would find the narrow, winding
corridor impassable. The passage would have to be
cleared of its wormed wood, and, moreover, greatly en-
larged. It would be easier to attack the untouched
timber and dig straight ahead. Is the insect capable of
doing so? I determined to find out.
I made some cavities of suitable size in some oak logs
that had been chopped in two, and in each of these cells
I placed a Capricorn that had just been transformed
from the grub. I then joined the two sides of the logs,
fastening them together with wire. When June came
I heard a sound of scraping inside the logs, and waited
anxiously to see if the Capricorns would appear. They
had hardly three-quarters of an inch to pierce. Yet not
one came out. On opening the logs I found all my cap-
tives dead. A pinch of sawdust represented all they
had done.
I had expected more from their sturdy mandibles.
In spite of their boring-tools the hermits died for lack of
[218]
THE CAPRICORN
skill. I tried enclosing some in reed-stumps, but even
this comparatively easy work was too much for them.
Some freed themselves, but others failed.
Notwithstanding his stalwart appearance the Capri-
corn cannot leave the tree-trunk by his own unaided
efforts. The truth is that his way is prepared for him by
the grub — that bit of intestine.
Some presentiment — to us an unfathomable mystery
I — causes the Capricorn-grub to leave its peaceful strong-
hold in the very heart of the oak and wriggle towards
the outside, where its foe the Woodpecker is quite likely
to gobble it up. At the risk of its life it stubbornly digs
and gnaws to the very bark. It leaves only the thinnest
film, the slenderest screen, between itself and the world
at large. Sometimes, even, the rash one opens the door-
way wide.
This is the Capricorn's way out. The insect has but
to file the screen a little with his mandibles, to bump
against it with his forehead, in order to bring it down.
He will even have nothing at all to do when the door-
way is open, as often happens. The unskilled car-
penter, burdened with his extravagant head-dress, will
come out from the darkness through this opening when
the summer heat arrives.
As soon as the grub has attended to the important
business of making a doorway into the world, it begins
to busy itself with its transformation into a Beetle.
[219]
FABRE'S BOOK OF INSECTS
First, it requires space for the purpose. So it retreats
some distance down its gallery, and in the side of the
passage digs itself a transformation-chamber more sump-
tuously furnished and barricaded than any I have ever
seen. It is a roomy hollow with curved walls, three to
four inches in length and wider than it is high. The
width of the cell gives the insect a certain degree of
freedom of movement when the time comes for forcing
the barricade, which is more than a close-fitting case
would do.
The barricade — a door which the larva builds as a
protection from danger — is twofold, and often three-
fold. Outside, it is a stack of woody refuse, of particles
of chopped timber; inside, a mineral lid, a concave cover,
all in one piece, of a chalky white. Pretty often, but
not always, there is added to these two layers an inner
casing of shavings.
Behind this threefold door the larva makes its arrange-
ments for its transformation. The sides of the chamber
are scraped, thus providing a sort of down formed of
ravelled woody fibres, broken into tiny shreds. This
velvety stuff is fixed on the wall, in a thick coating, as
fast as it is made. The chamber is thus padded through-
out with a fine swan's-down, a delicate precaution taken
by the rough grub out of kindness for the tender creature
it will become when it has cast its skin.
Let us now go back to the most curious part of the
[220]
THE CAPRICORN
furnishing, the cover or inner door of the entrance. It
is like an oval skull-cap, white and hard as chalk, smooth
within and rough without, with some resemblance to
an acorn-cup. The rough knots show that the material
is supplied in small, pasty mouthfuls, which become solid
outside in little lumps. The animal does not remove
them, because it is unable to get at them; but the inside
surface is polished, being within the grub's reach. This
singular lid is as hard and brittle as a flake of limestone.
It is, as a matter of fact, composed solely of carbonate
of lime, and a sort of cement which gives consistency to
the chalky paste.
I am convinced that this stony deposit comes from a
particular part of the grub's stomach, called the chylific
ventricle. The chalk is kept separate from the food,
and is held in reserve until the right time comes to dis-
charge it. This freestone factory causes me no astonish-
ment. It serves for various chemical works in different
grubs when undergoing transformation. Certain Oil-
beetles keep refuse in it, and several kinds of Wasps use
it to manufacture the shellac with which they varnish the
silk of their cocoons.
When the exit way is prepared, and the cell uphol-
stered in velvet and closed with a threefold barricade,
the industrious grub has finished its task. It lays aside
its tools, sheds its skin, and becomes a pupa — weakness
personified, in the swaddling-clothes of a cocoon. The
[221]
FABRE'S BOOK OF INSECTS
head is always turned towards the door. This is a
trifling detail in appearance; but in reality it is every-
thing. To lie this way or that in the long cell is a matter
of great indifference to the grub, which is very supple,
turning easily in its narrow lodging and adopting what-
ever position it pleases. The coming Capricorn will not
enjoy the same privileges. Stiffly encased in his horny
armour, he will not be able to turn from end to end ; he
will not even be capable of bending, if some sudden
curve should make the passage difficult. He must,
without fail, find the door in front of him, or he
will perish in the transformation-room. If the grub
should forget this little matter, and lie down to sleep
with its head at the back of the cell, the Capricorn would
be infallibly lost. His cradle would become a hopeless
dungeon.
But there is no fear of this danger. The "bit of in-
testine" knows too much about the future to neglect the
formality of keeping its head at the door. At the end
of spring the Capricorn, now in possession of his full
strength, dreams of the joys of the sun, of the festivals
of light. He wants to get out.
What does he find before him*? First, a heap of
filings easily dispersed with his claws; next, a stone lid
which he need not even break into fragments, for it comes
undone in one piece. It is removed from its frame with
a few pushes of the forehead, a few tugs of the claws.
[222]
THE CAPRICORN
In fact, I find the lid intact on the threshold of the
abandoned cell. Last comes a second mass of woody
remnants as easy to scatter as the first. The road is now
free : the Capricorn has but to follow the wide vestibule,
which will lead him, without any possibility of mistake,
to the outer exit. Should the doorway not be open, all
that he has to do is to gnaw through a thin screen, an
easy task. Behold him outside, his long antennae quiver-
ing with excitement.
What have we learnt from him? Nothing from him,
but much from his grub. This grub, so poor in organs
of sensation, gives us much to think about. It knows
that the coming Beetle will not be able to cut himself
a road through the oak, and it therefore opens one for
him at its own risk and peril. It knows that the Capri-
corn, in his stiff armour, will never be able to turn round
and make for the opening of the cell; and it takes care to
fall into its sleep of transformation with its head towards
the door. It knows how soft the pupa's flesh will be,
and it upholsters the bedroom with velvet. It knows
that the enemy is likely to break in during the slow work
of the transformation, and so, to make a protection
against attack, it stores lime inside its stomach. It
knows the future with a clear vision, or, to be accurate,
it behaves as if it knew the future.
What makes it act in this way? It is certainly not
taught by the experiences of its senses. What does it
[223]
FABRE'S BOOK OF INSECTS
know of the outside world? I repeat — as much as a
bit of intestine can know. And this senseless creature
astounds us I I regret that the philosopher Cbndillac,
instead of creating a statue that could smell a rose, did
not gift it with an instinct. How soon he would have
seen that the animals — including man — have powers
quite apart from the senses; inspirations that are born
with them, and are not the result of learning.
This curious life and this marvellous foresight are not
confined to one kind of grub. Besides the Capricorn of
the Oak there is the Capricorn of the Cherry-tree. In
appearance the latter is an exact copy of the former, on
a much smaller scale; but the little Capricorn has dif-
ferent tastes from its large kinsman's. If we search the
heart of the cherry-tree it does not show us a single 'grub
anywhere : the entire population lives between the bark
and the wood. This habit is only varied when trans-
formation is at hand. Then the grub of the cherry-tree
leaves the surface, and scoops out a cavity at a depth of
about two inches. Here the walls are bare: they are
not lined with the velvety fibres dear to the Cap-
ricorn of the Oak. The entrance is blocked, however,
by sawdust, and a chalky lid similar to the other except
in point of size. Need I add that the grub lies down
and goes to sleep with his head against the door? Not
one forgets to take this precaution.
There is also a Saperda of the Poplar and a Saperda
[224]
THE CAPRICORN
of the Cherry-tree. They have the same organisation
and the same tools ; but the former follows the methods
of the Capricorn of the Oak, while the latter imitates the
Capricorn of the Cherry-tree.
The poplar-tree is also inhabited by the Bronze Bu-
prestis, which takes no defensive measures before going
to sleep. It makes no barricade, no heap of shavings.
And in the apricot-tree the Nine-spotted Buprestis be-
haves in the same way. In this case the grub is inspired
by its intuitions to alter its plan of work to suit the
coming Beetle. The perfect insect is a cylinder; the
grub is a strap, a ribbon. The former, which wears
unyielding armour, needs a cylindrical passage; the
latter needs a very low tunnel, with a roof that it can
reach with the pads on its back. The grub therefore
changes its manner of boring: yesterday the gallery,
suited to a wandering life in the thickness of the wood,
was a wide burrow with a very low ceiling, almost a slot;
to-day the passage is cylindrical. A gimlet could not
bore it more accurately. This sudden change in the
system of roadmaking on behalf of the coming insect
once more shows us the foresight of this "bit of intes-
tine."
I could tell you of many other wood-eaters. Their
tools are the same; yet each species displays special
methods, tricks of the trade that have nothing to do with
the tools. These grubs, then, like so many insects, show
[225]
FABRE'S BOOK OF INSECTS
us that instinct is not made by the tools, so to speak,
but that the same tools may be used in various ways.
To continue the subject would be monotonous. The
general rule stands out very clearly from these facts:
the wood-eating grubs prepare the path of deliverance
for the perfect insect, which will merely have to pass
a barricade of shavings or pierce a screen of bark. By
a curious reversal of the usual state of things, infancy
is here the season of energy, of strong tools, of stubborn
work; mature age is the season of leisure, of indus-
trial ignorance, of idle diversions, without trade or pro-
fession. The providence of the human infant is the
mother; here the baby grub is the mother's providence.
With its patient tooth, which neither the peril of the
outside world nor the difficult task of boring through
hard wood is able to discourage, it clears away for her
to the supreme delights of the sun.
[226]
CHAPTER XV
LOCUSTS
THEIR VALUE
MIND you're ready, children, to-morrow morn-
ing before the sun gets too hot. We're
going Locust-hunting."
This announcement throws the household into great
excitement at bed-time. What do my little helpers see
in their dreams'? Blue wings, red wings, suddenly
flung out like fans; long saw-toothed legs, pale blue or
pink, which kick out when we hold their owners in our
fingers; great shanks that act like springs, and make
the insect leap forward as though shot from a catapult.
If there be one peaceful and safe form of hunting,
one in which both old age and childhood can share, it
is Locust-hunting. What delicious mornings we owe to
it I How delightful, when the mulberries are ripe, to
pick them from the bushes I What excursions we have
had, on the slopes covered with thin, tough grass, burnt
yellow by the sun I I have vivid memories of such
mornings, and my children will have them too.
[227]
FABRE'S BOOK OF INSECTS
Little Paul has nimble legs, a ready hand, and a
piercing eye. He inspects the clumps of everlastings,
and peers closely into the bushes. Suddenly a big Grey
Locust flies out like a little bird. The hunter first makes
off at full speed, then stops and gazes in wonder at this
mock Swallow flying far away. He will have better
luck another time. We shall not go home without a few
of those magnificent prizes.
Marie Pauline, who is younger than her brother,
watches patiently for the Italian Locust, with his pink
wings and carmine hind-legs; but she really prefers an-
other, the most ornamented of them* all. Her favourite
wears a St. Andrew's cross on the small of his back, which
is marked by four white, slanting stripes. He wears,
too, patches of green, the colour of verdigris on bronze.
With her hand raised in the air, ready to swoop down,
she approaches very softly, stooping low. Whoosh!
That's done it I The treasure is quickly thrust head-
first into a paper funnel, and plunges with one bound
to the bottom of it.
One by one our boxes are filled. Before the heat be-
comes too great to bear we are in possession of a number
of specimens. Imprisoned in my cages, perhaps they
will teach us something. In any case the Locusts have
given pleasure to three people at a small cost.
Locusts have a bad reputation, I know. The text-
books describe them as noxious. I take the liberty of
[228]
LOCUSTS
doubting whether they deserve this reproach, except, of
course, in the case of the terrible ravagers who are the
scourge of Africa and the East. Their ill repute has
been fastened on all Locusts, though they are, I consider,
more useful than harmful. As far as I know, our
peasants have never complained of them. What dam-
age do they do*?
They nibble the tops of the tough grasses which the
Sheep refuses to touch; they prefer the thin, poor grass
to the fat pastures; they browse on barren land that can
support none but them; they live on food that no stomach
but theirs could use.
Besides, by the time they frequent the fields the green
wheat — the only thing that might tempt them — has
long ago yielded its grain and disappeared. If they
happen to get into the kitchen-gardens and take a few
bites, it is not a crime. A man can console himself for
a piece bitten out of a leaf or two of salad.
To measure the importance of things by one's own
turnip-patch is a horrible method. The short-sighted
man would upset the order of the universe rather than
sacrifice a dozen plums. If he thinks of the insect at all,
it is only to kill it.
And yet, think what the consequences would be if all
the Locusts were killed. In September and October the
Turkeys are driven into the stubble, under charge of a
child armed with two long reeds. The expanse over
[229]
FABRE'S BOOK OF INSECTS
which the gobbling flock slowly spreads is bare, dry, and
burnt by the sun. At the most, a few ragged thistles
raise their heads. What do the birds do in this famine-
stricken desert? They cram themselves, that they may
do honour to the Christmas table; they wax fat; their
flesh becomes firm and good to eat. And pray, what do
they cram themselves with? With Locusts. They
snap them up, one here one there, till their greedy crops
are filled with the delicious stuffing, which costs nothing,
though its rich flavour will greatly improve the Christ-
mas Turkey.
When the Guinea-fowl roams about the farm, uttering
her rasping cry, what is it she seeks? Seeds, no doubt;
but above all Locusts, which puff her out under the wings
with a pad of fat, and give a better flavour to her flesh.
The Hen, too, much to our advantage, is just as fond of
them. She well knows the virtues of that dainty dish,
which acts as a tonic and makes her lay more eggs.
When left at liberty she rarely fails to lead her family
to the stubble-fields, so that they may learn to snap up
the nice mouthful skilfully. In fact, every bird in the
poultry-yard finds the Locust a valuable addition to his
bill of fare.
It is still more important outside the poultry-yard.
Any who is a sportsman, and knows the value of the Red-
legged Patridge, the glory of our southern hills, should
open the crop of the bird he has just shot. He will find
[230]
LOCUSTS
it, nine times out of ten, more or less crammed with
Locusts. The Partridge dotes on them, preferring them
to seeds as long as he can catch them. This highly-
flavoured, nourishing fare would almost make him forget
the existence of seeds, if it were only there all the year
round.
The Wheat-ear, too, who is so good to eat, prefers the
Locust to any other food. And all the little birds of
passage which, when autumn comes, call a halt in
Provence before their great pilgrimage, fatten them-
selves with Locusts as a preparation for the journey.
Nor does man himself scorn them. An Arab author
tells us :
"Grasshoppers" — (he means Locusts) — ''are of good
nourishment for men and Camels. Their claws, wings,
and head are taken away, and they are eaten fresh or
dried, either roast or boiled, and served with flesh, flour,
and herbs.
". . . Camels eat them greedily, and are given them
dried or roast, heaped in a hollow between two layers of
charcoal. Thus also do the Nubians eat them. . . .
"Once, when the Caliph Omar was asked if it were
lawful to eat Grasshoppers, he made answer:
" 'Would that I had a basket of them to eat.' "
"Wherefore, from this testimony, it is very sure that,
by the Grace of God, Grasshoppers were given to man
for his nourishment."
[231]
FABRE'S BOOK OF INSECTS
Without going as far as the Arab I feel prepared to say
that the Locust is a gift of God to a multitude of birds.
Reptiles also hold him in esteem. I have found him in
the stomach of the Eyed Lizard, and have often caught
the little Grey Lizard of the walls in the act of carrying
him off.
Even the fish revel in him, when good fortune brings
him to them. The Locust leaps blindly, and without
definite aim : he comes down wherever he is shot by the
springs in his legs. If the place where he falls happens
to be water, a fish gobbles him up at once. Anglers
sometimes bait their hooks with a specially attractive
Locust.
As for his being fit nourishment for man, except in the
form of Partridge and young Turkey, I am a little
doubtful. Omar, the mighty Caliph who destroyed the
library of Alexandria, wished for a basket of Locusts, it
is true, but his digestion was evidently better than his
brains. Long before his day St. John the Baptist lived
in the desert on Locusts and wild honey; but in his case
they were not eaten because they were good.
Wild honey from the pots of the Mason-bees is very
agreeable food, I know. Wishing to taste the Locust
also I once caught some, and had them cooked as the Arab
author advised. We all of us, big and little, tried the
queer dish at dinner. It was much nicer than the Cicadae
praised by Aristotle. I would go to the length of saying
[232]
LOCUSTS
it is good — without, however, feeling any desire for
more.
II
THEIR MUSICAL TALENT
The Locust possesses musical powers wherewith to
express his joys. Consider him at rest, blissfully digest-
ing his meal and enjoying the sunshine. With sharp
strokes of the bow, three or four times repeated with a
pause between, he plays his tune. He scrapes his sides
with his great hind-legs, using now one, now the other,
and now both at a time.
The result is very poor, so slight indeed that I am
obliged to make use of little Paul's sharp ear to make
sure that there is a sound at all. Such as it is, it is like
the squeaking of a needle-point pushed across a sheet of
paper. Their you have the whole song, which is very
nearly silence.
We can expect no more than this from the Locust's
very unfinished instrument. There is nothing here like
the Cricket's toothed bow and sounding-board. The
lower edge of the wing-cases is rubbed by the thighs, but
though both wing-cases and thighs are powerful they have
no roughnesses to supply friction, and there is no sign of
teeth.
This artless attempt at a musical instrument can pro-
[233]
FABRE'S BOOK OF INSECTS
duce no more sound than a dry membrane will emit when
you rub it yourself. And for the sake of this small result
the insect lifts and lowers its thigh in sharp jerks, and
appears perfectly satisfied. It rubs its sides very much
as we rub our hands together in sign of contentment,
with no intention of making a sound. That is its own
particular way of expressing its joy in life.
Observe the Locust when the sky is partly covered with
clouds, and the sun shines only at times. There comes a
rift in the clouds. At once the thighs begin to scrape, be-
coming more and more active as the sun grows hotter.
The strains are brief, but they are repeated as long as the
sunshine continues. The sky becomes overcast. Then
and there the song ceases; but is renewed with the next
gleam of sunlight, always in brief outburst. There is no
mistaking it : here, in these fond lovers of the light, we
have a mere expression of happiness. The Locust has his
moments of gaiety when his crop is full and the sun is
kind.
Not all the Locusts indulge in this joyous rubbing.
The Tryxalis, who has a pair of immensely long hind-
legs, keeps up a gloomy silence when even the sunshine is
brightest. I have never seen him move his shanks like
a bow ; he seem? unable to use them — so long are they —
for anything but hopping.
The big Grey Locust, who often visits me in the en-
closure, even in the depth of winter, is also dumb in
[234]
LOCUSTS
consequence of the excessive length of his legs. But he
has a peculiar way of diverting himself. In calm
weather, when the sun is hot, I surprise him in the rose-
mary bushes with his wings unfurled and fluttering
rapidly, as though for flight. He keeps up this per-
formance for a quarter of an hour at a time. His flutter-
ing is so gentle, in spite of its extreme speed, that it
creates hardly any rustling sound.
Others are still worse off. One of these is the
Pedestrian Locust, who strolls on foot on the ridges of
the Ventoux amid sheets of Alpine flowers, silvery,
white, and rosy. His colouring is as fresh as that of the
flowers. The sunlight, which is clearer on those heights
than it is below, has made him a costume combining
beauty with simplicity. His body is pale brown above
and yellow below, his big thighs are coral red, his hind-
legs a glorious azure-blue, with an ivory anklet in front.
But in spite of being such a dandy he wears too short a
coat.
His wing-cases are merely wrinkled slips, and his
wings no more than stumps. He is hardly covered as far
as the waist. Any one seeing him for the first time takes
him for a larva, but he is indeed the full-grown insect,
and he will wear this incomplete garment to the end.
With this skimpy jacket of course, music is impossible
to him. The big thighs are there; but there are no
wing-cases, no grating edge for the bow to rub upon.
[235]
FABRE'S BOOK OF INSECTS
The other Locusts cannot be described as noisy, but this
one is absolutely dumb. In vain have the most delicate
ears listened with all their might. This silent one must
have other means of expressing his joys. What they
are I do not know.
Nor do I know why the insect remains wthout wings,
a plodding wayfarer, when his near kinsmen on the same
Alpine slopes have excellent means of flying. He
possesses the beginnings of wings and wing-cases, gifts
inherited by the larva; but he does not develop these
beginnings and make use of them. He persists in
hopping, with no further ambition: he is satisfied to go
on foot, to remain -a Pedestrian Locust, when he might,
one would think, acquire wings. To flit rapidly from
crest to crest, over valleys deep in snow, to fly from one
pasture to another, would certainly be great advantages
to him. His fellow-dwellers on the mountain-tops
possess wings and are all the better for them. It would
be very profitable to extract from their sheaths the sails
he keeps packed away in useless stumps; and he does
not do it. Why?
No one knows why. Anatomy has these puzzles, these
surprises, these sudden leaps, which defy our curiosity.
In the presence of such profound problems the best thing
is to bow in all humility, and pass on.
[236]
LOCUSTS
III
THEIR EARLY DAYS
The Locust mother is not, in all cases, a model of
affection. The Italian Locust, having laboriously half-
buried herself in the sand, lays her eggs there and
immediately bounds away. She gives not a look at the
eggs, nor makes the least attempt to cover the hole where
they lie. It closes of its own accord, as best it can, by the
natural falling-in of the sand. It is an extremely casual
performance, marked by an utter absence of maternal
care.
Others do not forsake their eggs so recklessly. The
ordinary Locust with the blue-and-black wings, for in-
stance, after leaving her eggs in the sand, lifts her hind-
legs high, sweeps some sand into the hole, and presses it
down by stamping it rapidly. It is a pretty sight to
watch the swift action of her slender legs, giving alter-
nate kicks to the opening they are plugging. With
this lively trampling the entrance to the home is closed
and hidden away. The hole that contains the eggs
completely disappears, so that no ill-intentioned creature
could find it by sight alone.
Nor is this all. The power that works the two
rammers lies in the hinder thighs, which, as they rise and
[237]
FABRE'S BOOK OF INSECTS
fall, scrape lightly against the edge of the wing-cases.
This scraping produces a faint sound, similar to that
with which the insect placidly lulls itself to sleep in the
sun.
The Hen salutes with a song of gladness the egg she
just laid; she announces her performance to the whole
neighbourhood. The Locust celebrates the same event
with her thin scraper. "I have buried underground,"
she says, "the treasure of the future."
Having made the nest safe she leaves the spot,
refreshes herself after her exertions with a few mouth-
fuls of green stuif, and prepares to begin again.
The Grey Locust mother is armed at the tip of her
body — and so are other female Locusts in varying
degrees — with four short tools, arranged in pairs and
shaped like a hooked fingernail. On the upper pair,
which are larger than the others, these hooks are turned
upwards; on the lower and smaller pair they are turned
downwards. They form a sort of claw, and are scooped
out slightly, like a spoon. These are the pick-axes, the
boring-tools with which the Grey Locust works. With
these she bites into the soil, lifting the dry earth a little,
as quietly as if she were digging in soft mould. She
might be working in butter; and yet what the bore digs
into is hard, unyielding earth.
The best site for laying the eggs is not always found
at the first attempt. I have seen the mother make five
[238]
r
LOCUSTS
wells one after the other before finding a suitable place.
When at last the business is over, and the insect begins
to rise from the hole in which she is partly buried, one
can see that she is covering her eggs with milk-white
foam, similar to that of the Mantis.
This foamy matter often forms a button at the en-
trance to the well, a knot which stands up and attracts
the eye by its whiteness against the grey background of
the soil. It is soft and sticky, but hardens pretty soon.
When this closing button is finished the mother moves
away and troubles no more about her eggs, of which she
lays a fresh batch elsewhere after a few days.
Sometimes the foamy paste does not reach the sur-
face; it stops some way down, and before long is covered
with the sand that slips from the edge. But in the case
of my Locusts in captivity I always know, even when
it is concealed, exactly where the barrel of eggs lies.
Its structure is always the same, though there are varia-
tions in detail. It is always a sheath of solidified foam.
Inside, there is nothing but foam and eggs. The eggs
all lie in the lower portion, packed one on top of
another; and the upper part consists only of soft, yield-
ing foam. This portion plays an important part when
the young larvae are hatched. I will call it the ascend-
ing-shaft.
The wonderful egg-casket of the Mantis is not the
result of any special talent which the mother can ex-
[239]
FABRE'S BOOK OF INSECTS
ercise at will. It is due to mechanism. It happens
of itself. In the same way the Locusts have no in-
dustry of their own, especially devised for laying eggs
in a keg of froth. The foam is produced with the eggs,
and the arrangement of eggs at the bottom and centre,
and froth on the outside and the top, is purely
mechanical.
There are many Locusts whose egg-cases have to
last through the winter, since they do not open until
the fine weather returns. Though the soil is loose and
dusty at first, it becomes caked together by the winter
rains. Supposing that the hatching takes place a couple
of inches below the surface, how is this crust, this hard
ceiling, to be broken? How is the larva to come up
from below? The mother's unconscious art has ar-
ranged for that.
The young Locust finds above him, when he comes
out of the egg, not rough sand and hardened earth, but
a straight tunnel, with solid walls that keep all difficul-
ties away. This ascending-shaft is full of foam, which
the larva can easily penetrate, and which will bring
him quite close to the surface. Here only a finger's-
breadth of serious work remains to be done.
The greater part of the journey, therefore, is ac-
complished without effort. Though the Locust's build-
ing is done quite mechanically, without the least in-
telligence, it is certainly singularly well devised.
[240]
LOCUSTS
The little creature has now to complete his deliv-
erance. On leaving his shell he is of a whitish colour,
clouded with light red. His progress is made by worm-
like movements; and, so that it may be as easy as pos-
sible, he is hatched, like the young Grasshopper, in a
temporary jacket which keeps his antennae and legs
closely fixed to his body. Like the White-faced
Decticus he keeps his boring-tool at his neck. Here
there is a kind of tumour that swells and subsides
alternately, and strikes the obstacle before it as regularly
as a piston. When I see this soft bladder trying to
overcome the hardness of the earth I come to the un-
happy creature's aid, and damp the layer of soil.
Even then the work is terribly hard. How it must
labour, the poor little thing, how it must persevere with
its throbbing head and writhing loins, before it can clear
a passage for itself! The wee mite's efforts show us
plainly that the journey to the light of day is an enor-
mous undertaking, in which the greater number would
die but for the help of the exit-tunnel, the mother's
work.
When the tiny insect reaches the surface at last, it
rests for a moment to recover from all that fatigue.
Then suddenly the blister swells and throbs, and the
temporary jacket splits. The rags are pushed back
by the hind-legs, which are the last to be stripped. The
[241]
FABRE'S BOOK OF INSECTS
thing is done : the creature is free, pale in colouring as
yet, but possessing its final form as a larva.
Immediately the hind-legs, hitherto stretched in a
straight line, fall into the correct position. The legs
fold under the great thighs, and the spring is ready to
work. It works. Little Locust makes his entrance into
the world, and hops for the first time. I offer him a
bit of lettuce the size of my fingernail. He refuses
it. Before taking nourishment he must first mature
and grow in the sun.
IV
THEIR FINAL CHANGE
I have just beheld a stirring sight: the last change
of a Locust, the full-grown insect emerging from his
larval skin. It is magnificent. The object of my
enthusiasm is the Grey Locust, the giant who is so
common on the vines at vintage-time, in September.
On account of his size — he is as long as my finger — he
is easier to observe than any other of his tribe. The
event took place in one of my cages.
The fat, ungraceful larva, a rough sketch of the per-
fect insect, is usually pale green; but some are blue-
green, dirty yellow, red-brown, or even ashen-grey, like
the grey of the full-grown Locust. The hind-legs,
which are as powerful as those of mature age, have a
[242]
LOCUSTS
great haunch striped with red and a long shank shaped
like a two-edged saw.
The wing-cases are at present two skimpy, trian-
gular pinions, of which the free ends stand up like
pointed gables. These two coat-tails, of which the
material seems to have been clipped short with ridiculous
meanness, just cover the creature's nakedness at the
small of the back, and shelter two lean strips, the germs
of the wings. In brief, the sumptuous slender sails
of the near future are at present sheer rags, of such
meagre size as to be grotesque. From these miserable
envelopes there will come a marvel of stately elegance.
The first thing to be done is to burst the old tunic.
All along the corselet of the insect there is a line that
is weaker than the rest of the skin. Waves of blood
can be seen throbbing within, rising and falling alter-
nately, distending the skin until at last it splits at the
line of least resistance, and opens as though the two
symmetrical halves had been soldered. The split is
continued some little way back, and runs between the
fastenings of the wings: it goes up the head as far as
the base of the antennae, where it sends a short branch
to right and left.
Through this break the back is seen, quite soft, pale,
hardly tinged with grey. Slowly it swells into a larger
and larger hunch. At last it is wholly released. The
head follows, pulled out of its mask, which remains
[243]
FABRE'S BOOK OF INSECTS
in its place, intact in the smallest particular, but look-
ing strange with its great eyes that do not see. The
sheaths of the antennae, without a wrinkle, with nothing
out of order, and with their usual position unchanged,
hang over this dead face, which is now half transparent.
This means that the antennae within, although fitted
into narrow sheaths that enclose them as precisely as
gloves, are able to withdraw without disturbing the
covers in the smallest degree, or even wrinkling them.
The contents manage to slip out as easily as a smooth,
straight object could slip from a loose sheath. This
mechanism is even more remarkable in the case of the
hind-legs.
Now it is the turn of the fore-legs and intermediary
legs to shed their armlets and gauntlets, always without
the least rent, however small, without a crease of rumpled
material, or a trace of any change in the natural position.
The insect is now fixed to the top of the cage only by the
claws of the long hind-legs. It hangs perpendicularly
by four tiny hooks, head downwards, and it swings like a
pendulum if I touch the wire-gauze.
The wing-cases and wings now emerge. These are
four narrow strips, faintly grooved and looking like bits
of paper ribbon. At this stage they are scarcely a quarter
of their final length. They are so limp that they bend
under their own weight and sprawl along the insect's sides
in the wrong direction, with their points towards the head
[244]
LOCUSTS
of the Locust. Imagine four blades of thick grass, bent
and battered by a rain-storm, and you will have a fair
picture of the pitiable bunch formed by the future wings.
The hind-legs are next released. The great thighs
appear, tinted on their inner surface with pale pink,
which will soon turn into a streak of bright crimson.
They come out of the sheath quite easily, for the thick
haunch makes way for the tapering knuckle.
The shank is a different matter. The shank of the
full-grown insect bristles throughout its length with a
double row of hard, pointed spikes. Moreover, the lower
extremity ends in four large spurs. It is a genuine saw,
but with two parallel sets of teeth.
Now this awkwardly shaped skin is enclosed in a sheath
that is formed in exactly the same way. Each spur is
fitted into a similar spur, each tooth into the hollow of a
similar tooth. And the sheath is as close and as thin as a
coat of varnish.
Nevertheless the saw-like skin slips out of its long
narrow case without catching in it at any point whatever.
If I had not seen this happen over and over again I could
never have believed it. The saw does no injury to the
dainty scabbard which a puff of my breath is enough to
tear; the formidable rake slips through without leaving
the least scratch behind it.
One would expect that, because of the spiked armour,
the envelope of the leg would strip off in scales coming
[245]
FABRE'S BOOK OF INSECTS
loose of themselves, or would be rubbed off like dead
skin. But the reality exceeds all possible expectation.
From the spurs and spikes of the infinitely thin envelope
there are drawn spurs and spikes so strong that they
can cut soft wood. This is done without violence, the
discarded skin remains where it was, hanging by the
claws to the top of the cage, uncreased and untorn. The
magnifying-glass shows not a trace of rough usage.
If it were suggested that one should draw out a saw
from some sort of gold-beater's skin sheath which had
been exactly moulded on the steel, and that one should
perform the operation without making the least tear,
one would simply laugh. The thing would be im-
possible. Yet Nature makes light of such im-
possibilities; she can realise the absurd, in case of need.
The difficulty is overcome in this way. While the leg
is being liberated it is not rigid, as it will presently be.
It is soft and highly flexible. Where it is exposed to
view I see it bending and curving: it is as supple as
elastic cord. And farther on, where it is hidden, it is
certainly still softer, it is almost fluid. The teeth of the
saw are there, but have none of their future sharpness.
The spikes lie backwards when the leg is about to be
drawn back: as it emerges they stand up and become
solid. A few minutes later the leg has attained the
proper state of stiffness.
And now the fine tunic is wrinkled and rumpled, and
[246]
LOCUSTS
pushed back along the body towards the tip. Except at
this point the Locust is bare. After a rest of twenty
minutes he makes a supreme effort; he raises himself as
he hangs, and grabs hold of his cast skin. Then he
climbs higher, and fixes himself to the wire of the cage
with his four front feet. He loosens the empty husk
with one last shake, and it falls to the ground. The
Locust's transformation is conducted in much the same
way as the Cicada's.
The insect is now standing erect, and therefore the
flexible wings are in the right position. They are no
longer curved backwards like the petals of a flower, they
are no longer upside down; but they still look shabby
and insignificant. All that we see is a few wrinkles, a
few winding furrows, which tell us that the stumps are
bundles of cunningly folded material, arranged so as
to take up as little space as possible.
Very gradually they expand, so gradually that their
unfolding cannot be seen even under the microscope.
The process continues for three hours. Then the wings
and wing-cases stand up on the Locust's back like a huge
set of sails, sometimes colourless, sometimes pale-green,
like the Cicada's wings at the beginning. One is amazed
at their size when one thinks of the paltry bundles that
represented them at first. How could so much stuff find
room there ^
The fairy tale tells us of a grain of hempseed that
[247]
FABRE'S BOOK OF INSECTS
contained the under-linen of a princess. Here is a grain
that is even more astonishing. The one in the story
took years and years to sprout and multiply, till at last
it yielded the hemp required for the trousseau: the
Locust's tiny bundle supplies a sumptuous set of sails
in three hours. They are formed of exquisitely fine
gauze, a network of innumerable tiny bars.
In the wing of the larva we can see only a few un-
certain outlines of the future lace-work. There is no-
thing to suggest the marvellous fabric whose every mesh
will have its form and place arranged for it, 'with
absolute exactness. Yet it is there, as the oak is inside
the acorn.
There must be something to make the matter of the
wing shape itself into a sheet of gauze, into a labyrinth
of meshes. There must be an original plan, an ideal
pattern which gives each atom its proper place. The
stones of our buildings are arranged in accordance with
the architect's plan; they form an imaginary building be-
fore they exist as a real one. In the same way a Locust's
wing, that sumptuous piece of lace emerging from a
miserable sheath, speaks to us of another Architect, the
Author of the plans which Nature must follow in her
labours.
[248]
CHAPTER XVI
THE ANTHRAX FLY
I
A STRANGE MEAL
I MADE the acquaintance of the Anthrax in 1855 at
Carpentras, when I was searching the slopes of
which I have already told you, the slopes beloved
of the Anthophora-bees. Her curious pupa, so power-
fully equipped to force an outlet for the perfect insect,
which is incapable of the least effort, seemed worthy of
investigation. For that pupa is armed with a plough-
share in front, a trident at its tail, and rows of harpoons
on its back, with which to rip open the Osmia-bee's cocoon
and break through the hard crust of the hill-side.
Let us, some day in July, knock away the pebbles that
fasten the nests of the Mason-bees to the sloping ground
on which they are built. Loosened by the shock, the
dome comes off cleanly, all in one piece. Moreover —
and this is a great advantage — the cells are all exposed
at the base of the nest, for at this point they have no other
wall than the surface of the pebble. Without any scrap-
ing, which would be wearisome work for us and danger-
ous to the Bees, we have all the cells before our eyes, to-
[249]
FABRE'S BOOK OF INSECTS
gether with their contents — a silky, amber-yellow cocoon,
as delicate and transparent as the skin of an onion. Let
us split the dainty wrappers with the scissors, cell by cell,
one after another. If fortune be at all kind, as it always
is to the persevering, we shall end by finding cocoons
harbouring two larvse together, one more or less faded
in appearance, the other fresh and plump. We shall also
find some, no less plentiful, in which the withered larva
is accompanied by a family of little grubs wriggling un-
easily round it.
It is easy to see that a tragedy is happening under the
cover of the cocoon. The flabby, faded larva is the
Mason-bee's. A month ago, in June, having finished
its ration of honey, it wove itself a silken sheath in which
to take the long sleep that precedes its transformation.
It was bulging with fat, and was a rich and a defenceless
morsel for any enemy that could reach it. And enemies
did reach it. In spite of obstacles that might well seem
insurmountable, the wall of mortar and dome-shaped
cover, the enemy grubs appeared in the secret retreat, and
began to eat the sleeper. Three different species take
part in this murderous work, often in the same nest, in
adjoining cells. We will concern ourselves only with
the Anthrax Fly.
The grub, when it has eaten its victim and is left alone
in the Mason-bee's cocoon, is a naked worm, smooth, leg-
less, and blind. It is creamy-white, and each of its
[250]
THE ANTHRAX FLY
segments or divisions forms a perfect ring, very much
curved when at rest, but almost straight when disturbed.
Including the head I can count thirteen segments, well-
marked in the middle of the body, but in the fore-part
difficult to distinguish. The white, soft head shows no
sign of any mouth, and is no bigger than a tiny pin's
head. The grub has four pale red stigmata, or openings
through which to breathe, two in front and two behind,
as is the rule among Flies. It has no walking-apparatus
whatever; it is absolutely incapable of shifting its
position. If I disturb its rest, it curves and straightens
itself alternately, tossing about violently where it lies;
but it does not manage to progress.
But the most interesting point about the grub of the
Anthrax is its manner of eating. A most unexpected
fact attracts our attention : the curious ease with which
this larva leaves and returns to the Bee-grub on which it
is feeding. After watching flesh-eating grubs at
hundreds and hundreds of meals, I suddenly find myself
confronted with a manner of eating that is entirely un-
like anything I ever saw before.
This, for instance, is the Amophila-grub's way of de-
vouring its caterpillar. A hole is made in the victim's
side, and the head and neck of the grub dives deep into
the wound. It never withdraws its head, never pauses
to take breath. The voracious animal always goes for-
ward, chewing, swallowing, digesting, until the cater-
[251]
FABRE'S BOOK OF INSECTS
pillar's skin is empty. Once the meal is begun, the
creature does not budge as long as the food lasts. If
moved by force it hesitates, and hunts about for the
exact spot where it left off eating; for if the caterpillar
be attacked at a fresh point it is liable to go bad.
In the case of the Anthrax-grub there is none of this
mangling, none of this persistent clinging to the original
wound. If I tease it with the tip of a pointed brush it at
once retires, and there is no wound to be seen on the
victim, no sign of broken skin. Soon the grub once more
applies its pimple-head to its meal, at any point, no
matter where, and keeps itself fixed there without any
effort. If I repeat the touch with the brush I see the
same sudden retreat and the same calm return to the
meal.
The ease with which this larva grips, leaves, and re-
grips its victim, now here, now there, and always with-
out a wound, shows that the rnouth of the Anthrax is not
armed with fangs that can dig into the skin and tear it.
If the flesh were gashed by pincers of any kind, one or
two attempts would be necessary before they could leave
go or take hold again; and besides, the skin would be
broken. There is nothing of the kind : the grub simply
glues its mouth to its prey, and withdraws it. It does
not chew its food like the other flesh-eating grub: it
does not eat, it inhales.
This remarkable fact led me to examine the mouth
[252]
THE ANTHRAX FLY
under the microscope. It is a small conical crater, with
yellowish-red sides and very faint lines running round
it. At the bottom of this funnel is the opening of the
throat. There is not the slightest trace of mandibles or
jaws, or any object capable of seizing and grinding food.
There is nothing at all but the bowl-shaped opening. I
know of no other example of a mouth like this, which
I can only compare to a cupping-glass. Its attack is a
mere kiss, but what a cruel kiss I
To observe the working of this curious machine I
placed a new-born Anthrax-grub, together with its prey,
in a glass tube. Here I was able to watch the strange re-
past from beginning to end.
The Anthrax-grub — the Bee's uninvited guest — is
fixed by its mouth or sucker to any convenient part of
the plump Bee-grub. It is ready to break off its kiss
suddenly, should anything disturb it, and to resume it as
easily when it wishes. After three or four days of this
curious contact the Bee-grub, formerly so fat, glossy, and
healthy, begins to look withered. Her sides fall in, her
fresh colour fades, her skin becomes covered with little
folds, and she is evidently shrinking. A week is hardly
passed when these signs of exhaustion increase to a
startling degree. The victim is flabby and wrinkled, as
though borne down by her own weight. If I move her
from her place she flops and sprawls like a half-filled
indiarubber bottle. But the kiss of the Anthrax goes on
[253]
FABRE'S BOOK OF INSECTS
emptying her : soon she is but a sort of shrivelled bladder,
growing smaller and smaller from hour to hour. At
length, between the twelfth and fifteenth day, all that re-
mains of the Mason-bee's larva is a little white grain,
hardly as large as a pin's head.
If I soften this small remnant in water, and then blow
into it through a very fine glass tube, the skin fills out
and resumes the shape of the larva. There is no outlet
anywhere for the compressed air. It is intact : it is no-
where broken. This proves that, under the cupping-
glass of the Anthrax, the skin has been drained through
its pores.
The devouring grub, in making its attack, chooses its
moment very cunningly. It is but an atom. Its mother,
a feeble Fly, has done nothing to help it. She has no
weapons; and she is quite incapable of penetrating the
Mason-bee's fortress. The future meal of the Anthrax
has not been paralysed, nor injured in any way. The
parasite arrives — we shall presently see how; it arrives,
scarcely visible, and having made its preparations it in-
stalls itself upon its monstrous victim, whom it is going
to drain to the very husk. And the victim, though not
paralysed nor in any way lacking in vitality, lets it have
its way, and is sucked dry without a tremor or a quiver of
resistance. No corpse could show greater indifference
to a bite.
Had the Anthrax-grub appeared upon the scene
[254]
THE ANTHRAX FLY
earlier, when the Bee-grub was eating her store of honey,
things would surely have gone badly with it. The
victim, feeling herself bled to death by that ravenous
kiss, would have protested with much wriggling of body
and grinding of mandibles. The intruder would have
perished. But at the hour chosen so wisely by it all
danger is over. Enclosed in her silken sheath, the larva
is in the torpid state that precedes her transformation
into a Bee. Her condition is not death, but neither is it
life. So there is no sign of irritation when I stir her with
a needle, nor when the Anthrax-grub attacks her.
There is another marvellous point about the meal of
the Anthrax-grub. The Bee-grub remains alive until the
very end. Were she really dead it would, in less than
twenty-four hours, turn a dirty-brown colour and de-
compose. But during the whole fortnight that the meal
lasts, the butter-colour of the victim continues unaltered,
and there is no sign of putrefaction. Life persists un-
til the body is reduced to nothing. And yet, if I myself
give her a wound, the whole body turns brown and soon
begins to rot. The prick of a needle makes her decom-
pose. A mere nothing kills it; the atrocious draining of
its strength does not.
The only explanation I can suggest is this, and it is no
more than a suggestion. Nothing but fluids can be
drawn by the sucker of the Anthrax through the unpierced
skin of the Bee-grub : no part of the breathing-apparatus
FABRE'S BOOK OF INSECTS
or the nervous system can pass. As these two essentials
remain uninjured, life goes on until the fluid contents of
the skin are entirely exhausted. On the other hand, if
I myself injure the larva of the Bee, I disturb the nervous
or the air-conducting system, and the bruised part
spreads a taint all over the body.
Liberty is a noble possession, even in an insignificant
grub; but it has its dangers everywhere. The Anthrax
escapes these dangers only on the condition of being, so
to speak, muzzled. It finds its own way into the Bee's
dwelling, quite independently of its mother. Unlike
most of the other flesh-eating larvse it is not fixed by its
mother's care at the most suitable spot for its meal. It
is perfectly free to attack its prey where it chooses. If it
had a set of carving-tools, of jaws and mandibles, it
would meet with a speedy death. It would split open
its victim and bite it at random, and its food would rot.
Its freedom of action would kill it.
II
THE WAY OUT
There are other grub-eaters which drain their victims
without wounding them, but not one, among those I
know, reaches such perfection in this art as the Anthrax-
grub. Nor can any be compared with the Anthrax as re-
gards the means brought into play in order to leave the
[256]
THE ANTHRAX FLY
cell. The others, when they become perfect insects, have
implements for mining and demolishing. They have
stout mandibles, capable of digging the ground, of
pulling down clay partition-walls, and even of grinding
the Mason-bee's tough cement to powder. The Anthrax,
in her final form, has nothing like this. Her mouth is
a short, soft proboscis, good at most for soberly licking
the sugary fluid from the flowers. Her slim legs are so
feeble that to move a grain of sand would be too heavy a
task for them, enough to strain every joint. Her great
stiff wings, which must remain full-spread, do not allow
her to slip through a narrow passage. Her delicate suit
of downy velvet, from which you take the bloom by
merely breathing on it, could not withstand the contact
of rough tunnels. She is unable to enter the Mason-
bee's cells to lay her egg, and equally unable to leave it
when the time comes to free herself and appear in broad
daylight.
And the grub, for its part, is powerless to prepare the
way for the coming flight. That buttery little cylinder,
owning no tools but a sucker so flimsy and small that it
is barely visible through the magnifying-glass, is even
weaker than the full-grown insect, which at least flies
and walks. The Mason-bee's cell seems to this creature
like a granite cave. How can it get out ? The problems
would be insoluble to these two incapables, if nothing
else played its part.
[257]
FABRE'S BOOK OF INSECTS
Among insects the pupa — the transition stage, when
the creature is no longer a grub but is not yet a perfect
insect — is generally a striking picture of complete weak-
ness. A sort of mummy, tightly bound in swaddling-
clothes, motionless and unconscious, it awaits its trans-
formation. Its tender flesh is hardly solid; its limbs are
transparent as crystals, and are held fixed in their place,
lest a movement should disturb the work of development.
In the same way, to secure his recovery, a patient whose
bones are broken is held bound in the surgeon's bandages.
Well, here, by a strange reversal of the usual state of
things, a stupendous task is laid upon the pupa of the
Anthrax. It is the pupa that has to toil, to strive, to ex-
haust itself in efforts to burst the wall and open the way
out. To the pupa falls the desperate duty, to the full-
grown insect the joy of resting in the sun. The result of
these unusual conditions is that the pupa possesses a
strange and complicated set of tools that is in no way
suggested by the grub nor recalled by the perfect Fly.
This set of tools includes a collection of ploughshares,
gimlets, hooks, spears, and other implements that are not
found in our trades nor named in our dictionaries. I
will do my best to describe the strange gear.
By the time that July is nearly over the Anthrax has
finished eating the Bee-grub. From that time until the
following May it lies motionless in the Mason-bee's
cocoon, beside the remains of its victim. When the fine
[258]
THE ANTHRAX FLY
days of May arrive it shrivels, and casts its skin; and it is
then that the pupa appears, fully clad in a stout, reddish,
horny hide.
The head is round and large, and is crowned on top and
in front with a sort of diadem of six hard, sharp, black
spikes, arranged in semi-circle. This sixfold plough-
share is the chief digging-implement. Lower down the
instrument is finished off with a separate group of two
small black spikes, placed close together.
Four segments in the middle of the body are armed on
the back with a belt of little horny arches, set in the skin
upside down. They are arranged parallel to one an-
other, and are finished at both ends with a hard, black
point. The belt forms a double row of little thorns,
with a hollow in between. There are about two
hundred spikes on the four segments. The use of this
rasp, or grater, is obvious : it helps the pupa to steady it-
self on the wall of the gallery as the work proceeds.
Thus anchored on a host of points the brave pioneer is
able to hit the obstacle harder with its crown of awls.
Moreover, to make it more difficult for the instrument to
recoil, there are long, stiff bristles, pointing backwards,
scattered here and there among the rows of spikes.
There are some also on other segments, and on the sides
they are arranged in clusters. Two more belts of thorns,
less powerful than the others, and a sheaf of eight spikes
at the tip of the body — two of which are longer than the
[259]
FABRE'S BOOK OF INSECTS
rest — completes the strange boring-machine that pre-
pares an outlet for the feeble Anthrax.
About the end of May the colouring of the pupa alters,
and shows that the transformation is close at hand. The
head and fore-part of the creature become a handsome,
shiny black, prophetic of the black livery worn by the
coming insect. I was anxious to see the boring-tools in
action, and, since this could not be done in natural con-
ditions, I confined the Anthrax in a glass tube, between
two thick stoppers of sorghum-pith. The space between
the stoppers was about the same size as the Bee's cell,
and the partitions, though not so strong as the Bee's
masonry, were firm enough to withstand considerable
effort. On the other hand the side-walls, being of glass,
could not be gripped by the toothed belts, which made
matters much harder for the worker.
No matter: in the space of a single day the pupa
pierced the front partition, three-quarters of an inch
thick. I saw it fixing its double ploughshare against the
back partition, arching itself into a bow, and then
suddenly releasing itself and striking the stopper in front
of it with its barbed forehead. Under the blows of the
spikes the pith slowly crumbled to pieces, atom by atom.
At long intervals the method of work changed. The
animal drove its crown of awls into the pith, and fidgeted
and swayed about for a time ; then the blows began again.
[260]
THE ANTHRAX FLY
Now and then there were intervals of rest. At last the
hole was made. The pupa slipped into it, but did not
pass through entirely. The head and chest appeared be-
yond the hole, but the rest of the body remained held in
the tunnel.
The glass cell certainly puzzled my Anthrax. The
hole through the pith was wide and irregular: it was a
clumsy breach and not a gallery. When made through
the Mason-bee's walls it is fairly neat, and exactly of the
animal's diameter. For narrowness and evenness in the
exit-tunnel are necessary. The pupa always remains
half-caught in it, and even pretty securely fixed by the
graters on its back. Only the head and chest emerge
into the outer air. A fixed support is indispensable, for
without it the Anthrax could not issue from her horny
sheath, unfurling her great wings and drawing out her
slender legs.
She therefore remains steadily fixed by the graters on
her back, in the narrow exit-gallery. All is now ready.
The transformation begins. Two slits appear on the
head: one along the forehead, and a second, crossing it,
dividing the skull in two and extending down the chest.
Through this cross-shaped opening the Anthrax Fly
suddenly appears. She steadies herself upon her trem-
bling legs, dries her wings and takes to flight, leaving her
cast skin at the doorway of the gallery. The sad-
[261]
FABRE'S BOOK OF INSECTS
coloured Fly has five or six weeks before her wherein to
explore the clay nests amid the thyme and to take her
small share of the joys of life.
Ill
THE WAY IN
IF you have paid attention to this story of the Anthrax
Fly, you must have noticed that it is incomplete.
The Fox in the fable saw how the Lion's visitors en-
tered his den, but did not see how they went out. With
us the case is reversed: we know the way out of the
Mason-bee's fortress, but we do not know the way in. To
leave the cell whose owner it has eaten, the Anthrax be-
comes a boring-tool. When the exit-tunnel is opened
this tool splits like a pod bursting in the sun, and from
the strong framework there escapes a dainty Fly. A soft
bit of fluff that contrasts strangely with the roughness of
the prison whence it comes. On this point we know
pretty well what there is to know. But the entrance of
the grub into the cell puzzled me for a quarter of a cen-
tury.
It is plain that the mother cannot place her egg in the
Bee's cell, which is closed and barricaded with a cement
wall. To pierce it she would have to become a boring-
tool once more, and get into the cast-off rags which she
left at the doorway of the exit-tunnel. She would have
[262]
THE ANTHRAX FLY
to become a pupa again. For the full-grown Fly has no
claws, nor mandibles, nor any implement capable of
working its way through the wall.
Can it be, then, the grub that makes its own way into
the storeroom, that same grub that we have seen sucking
the life out of the Bee's larva'? Let us call the creature
to mind : a little oily sausage, which stretches and curls
up just where it lies, without being able to shift its
position. Its body is a smooth cylinder, its mouth a
circular lip. It has no means whatever of moving; not
even a hair or a wrinkle to enable it to crawl. It can do
nothing but digest its food. It is even less able than the
mother to make its way into the Mason-bee's dwelling.
And yet its provisions are there: they must be reached:
it is a matter of life and death. How does the Fly set
about it^ In the face of this puzzle I resolved to
attempt an almost impossible task and watch the
Anthrax from the moment it left the egg.
Since these Flies are not really plentiful in my own
neighbourhood I made an expedition to Carpentras, the
dear little town where I spent my twentieth year. The
old college where I made my first attempts as a teacher
was unchanged in appearance. It still looked like a
penitentiary. In my early days it was considered un-
wholesome for boys to be gay and active, so our system
of education applied the remedy of melancholy and
gloom. Our houses of instruction were above all houses
[263]
FABRE'S BOOK OF INSECTS
of correction. In a yard between four walls, a sort of
bear-pit, the boys fought to make room for their games
under a spreading plane-tree. All round it were cells
like horseboxes, without light or air: those were the
class-rooms.
I saw, too, the shop where I used to buy tobacco as I
came out of the college; and also my former dwelling,
now occupied by monks. There, in the embrasure of a
window, sheltered from profane hands, between the
closed outer shutters and the panes, I kept my chemicals —
bought for a few ,sous saved out of the housekeeping
money. My experiments, harmless or dangerous, were
made on a corner of the fire, beside the simmering broth.
How I should love to see that room again, where I pored
over mathematical problems ; and my familiar friend the
blackboard, which I hired for five francs a year, and
could never buy outright for want of the necessary cash I
But I must return to my insects. My visit to
Carpentras, unfortunately, was made too late in the
year to be very profitable. I saw only a few Anthrax
Flies hovering round the face of the cliff. Yet I did not
despair, because it was plain that these few were not
there to take exercise, but to settle their families.
So I took my stand at the foot of the rock, under a
broiling sun, and for half a day I followed the move-
ments of my Flies. They flitted quietly in front of the
slope, a few inches away from the earthly covering.
[264]
THE ANTHRAX FLY
They went from one Bee's nest to another, but without
attempting to enter. For that matter, the attempt would
be useless, for the galleries are too narrow to admit their
spreading wings. So they simply explore the cliff, going
to and fro, and up and down, with a flight that was now
sudden, now smooth and slow. From time to time I
saw one of them approach the wall and touch the earth
suddenly with the tip of her body. The proceeding
took no longer than the twinkling of an eye. When it
was over the insect rested a moment, and then resumed
flight.
I was certain that, at the moment when the Fly tapped
the earth, she laid her eggs on the spot. Yet, though I
rushed forward and examined the place with my lens, I
could see no egg. In spite of the closest attention I
could distinguish nothing. The truth is that my state
of exhaustion, together with the blinding light and
scorching heat, made it difficult for me to see anything.
Afterwards, when I made the acquaintance of the tiny
thing that comes out of that egg, my failure no longer sur-
prised me : for even in the leisure and peace of my study
I have the greatest difficulty in finding the infinitesimal
creature. How then could I see the egg, worn out as 1
was under the sun-baked cliff?
None the less I was convinced that I had seen the
Anthrax Flies strewing their eggs, one by one, on the
spots frequented by the Bees who suit their grubs. They
[265]
FABRE'S BOOK OF INSECTS
take no precaution to place the egg under cover, and in-
deed the structure of the mother makes any such pre-
caution impossible. The egg, that delicate object, is
laid roughly in the blazing sun, among grains of sand,
in some wrinkle of the chalk. It is the business of the
young grub to manage as best it can.
The next year I continued my investigations, this time
on the Anthrax of the Chalicodoma, a Bee that abounds
in my own neighbourhood. Every morning I took the
field at nine o'clock, when the sun begins to be unendur-
able. I was prepared to i^ome back with my head aching
from the glare, if only I could bring home the solution of
my puzzle. The greater the heat, the better my chances
of success. What gives me torture fills the insect with
delight; what prostrates me braces the Fly.
The road shimmers like a sheet of molten steel. From
the dusty, melancholy olive-trees rises a mighty, throb-
bing hum, the concert of the Cicadae, who sway and
rustle with increasing frenzy as the temperature in-
creases. The Cicada of the Ash adds its strident scrap-
ings to the single note of the Common Cicada. This is
the moment ! For five or six weeks, of tenest in the morn-
ing, sometimes in the afternoon, I set myself to explore
the rocky waste.
There were plenty of the nests I wanted, but I could
not see a single Anthrax on their surface. Not one
settled in front of me to lay her egg. At most, from time
[266]
THE ANTHRAX FLY
to time, I could see one passing far away, with an im-
petuous rush. I would lose her in the distance; and that
was all. It was impossible to be present at the laying
of the egg. In vain I enlisted the services of the small
boys who keep the sheep in our meadows, and talked
to them of a big black Fly and the nests on which
she ought to settle. By the end of August my last
illusions were dispelled. Not one of us had succeeded
in seeing the big black Fly perching on the dome of
the Mason-bee.
The reason is, I believe, that she never perches there.
She comes and goes in every direction across the stony
plain. Her practised eye can detect, as she flies, the
earthen dome which she is seeking, and having found it
she swoops down, leaves her egg on it, and makes off
without setting foot on the ground. Should she take a
rest it will be elsewhere, on the soil, on a stone, on
a tuft of lavender or thyme. It is no wonder that
neither I nor my young shepherds could find her egg.
Meanwhile I searched the Mason-bees' nests for grubs
just out of the egg. My shepherds procured me heaps
of the nests, enough to fill baskets and baskets; and
these I inspected at leisure on my work-table. I took
the cocoons from the cells, and examined them within
and without: my lens explored their innermost recesses,
the sleeping larva, and the walls. Nothing, nothing,
nothing I For a fortnight and more nests were searched
[267]
FABRE'S BOOK OF INSECTS
and rejected, and heaped up in a corner. My study
was crammed with them. In vain I ripped up the
cocoons; I found nothing. It needed the sturdiest faith
to make me persevere.
At last I saw, or seemed to see, something move on
the Bee's larva. Was it an illusion *? Was it a bit
of down stirred by my breath? It was not an illusion;
it was not a bit of down; it was really and truly a
grub I But at first I thought the discovery unimportant,
because I was so greatly puzzled by the little creature's
appearance.
In a couple of days I was the owner of ten such worms
and had placed each of them in a glass tube, together
with the Bee-grub on which it wriggled. It was so tiny
that the least fold of skin concealed it from my sight.
After watching it one day through the lens I sometimes
failed to find it again on the morrow. I would think
it was lost : then it would move, and become visible once
more.
For some time the belief had been growing in me that
the Anthrax had two larval forms, a first and a second,
the second being the form I knew, the grub we have al-
ready seen at its meals. Was this new discovery, I
asked myself, the first form? Time showed me that
it was. For at last I saw my little worms transform
themselves into the grub I have already described, and
make their first start at draining their victims with kisses.
[268]
THE ANTHRAX FLY
A few moments of satisfaction like those I then enjoyed
make up for many a weary hour.
This tiny worm, the first form or "primary larva" of
the Anthrax, is very active. It tramps over the fat
sides of its victim, walking all round it. It covers the
ground pretty quickly, buckling and unbuckling by
turns, very much after the manner of the Looper-cater-
pillar. Its two ends are its chief points of support.
When walking it swells out, and then looks like a bit
of knotted string. It has thirteen rings or segments,
including its tiny head, which bristles in front with
short, stiff hairs. There are four other pairs of bristles
on the lower surface, and with the help of these it walks.
For a fortnight the feeble grub remains in this
condition, without growing, and apparently without eat-
ing. Indeed, what could it eat? In the cocoon there
is nothing but the larva of the Mason-bee, and the worm
cannot eat this before it has the sucker or mouth that
comes with the second form. Nevertheless, as I said
before, though it does not eat it is far from idle. It ex-
plores its future dish, and runs all over the neighbor-
hood.
Now, there is a very good reason for this long fast.
In the natural state of the Anthrax-grub it is necessary.
The egg is laid by the mother on the surface of the nest,
at a distance from the Bee's larva, which is protected
by a thick rampart. It is the business of the new-born
[269]
FABRE'S BOOK OF INSECTS
grub to make its way to its provisions, not by violence,
of which it is incapable, but by patiently slipping
through a maze of cracks. It is a very difficult task,
even for this slender worm, for the Bee's masonry is
exceedingly compact. There are no chinks due to bad
building, no cracks due to the weather. I see but one
weak point, and that only in a few nests : it is the line
where the dome joins the surface of the stone. This
weakness so seldom occurs that I believe the Anthrax-
grub is able to find an entrance at any spot on the dome
of the Bee's nest.
The grub is extremely weak, and has nothing but
invincible patience. How long it takes to work its way
through the masonry I cannot say. The work is so
laborious and the worker so feeble I In some cases I
believe it may be months before the slow journey is ac-
complished. So it is very fortunate, you see, that this
first form of the Anthrax, which exists only in order to
pierce the walls of the Bees' nest, should be able to live
without food.
At last I saw my young worms shrink, and rid them-
selves of their outer skin. They then appeared as the
grub I knew and was so anxiously expecting, the grub
of the Anthrax, the cream-colored cylinder with the little
button of a head. Fastening its round sucker to the
Bee-grub, it began its meal. You know the rest.
Before taking leave of this tiny animal let us dwell
[270]
THE ANTHRAX FLY
for a moment on its marvellous instinct. Picture it as
having just left the egg, just awakened to life under
the fierce rays of the sun. The bare stone is its cradle;
there is no one to welcome it as it enters the world, a
mere thread of half-solid substance. Instantly it starts
on its struggle with the flint. Obstinately it sounds
each pore of the stone; it slips in, crawls on, retreats,
begins again. What inspiration urges it towards its
food, what compass guides it? What does it know of
those depths, or of what lies in them? Nothing. What
does the root of a plant know of the earth's fruitfulness?
Again, nothing. Yet both the root and the worm make
for the nourishing spot. Why? I do not understand.
I do not even try to understand. The question is far
above us.
We have now followed the complete history of the
Anthrax. Its life is divided into four periods, each of
which has its special form and its special work. The
primary larva enters the Bees' nest, which contains pro-
visions; the secondary larva eats those provisions; the
pupa brings the insect to light by boring through the
enclosing wall ; the perfect insect strews its eggs. Then
the story starts afresh.
[271]